Note: Descriptions are shown in the official language in which they were submitted.
7~
FIELD OF THE INVENTION
The present invention relates to a novel cast steel
products, and particularly to a novel continuously cast steel
product having a uniform distribution of constituents, and -the
method of producing said product.
BACKGROUND ART
A desirable property of steel in its as-cast condition
is a uniform distribution within the steel product of the con-
stituents and impurities normally found in steel. As used
herein, what applicants believe to be the meaning of these
terms is their standard meanings in the art, that is, "consti-
tuent" means one of the ingredients which make up a chemical
system or a phase or combination of phases which occur in a
characteristic configuration in an alloy microstructure, while
"impurities" means elements or com~ounds whose presence in any
material is undesired. Constituents, as used herein, then,
would include the materials combined into a chemical system to
produce the particular type of steel being ~ast but would not
include the impurities, or undesired elements or compounds
present in the cast metal. In any case, segregation of the
components in the cast steel makes it less suitable for subse-
~uent processing such as forging or rolling into rod and then
drawing into wire. As used herein, the term "sagregation"
also has what applicants believe to be its normal meaning in
the art, that is, segregation is a term used in refe~rence to
the non-uniform distribution or concentration of constituents
(or impurities) which arises during the solidification of the
metal. A concentration or accumulation of im~urities in
- various positions within the metal is, for example, referred
to in the art as segregation. The segregation that occurs
between the arms of dendrites is referred to as minor or
microsegregaton and thus the composition may vary within a
single crystel. Macrosegregation occurs around primary or
secondary shrinkage cavities, such as pipe and in similar
~,
;.~'
'73
regions, and is often revealed as marked lines, having a
pronounced erect or inverted cone shape, which are made evi-
dent whenthe ingots are sectioned and etched. Zones of segre-
gation tend to occur in the middle regions of the casting and
usually within that part mainly occupied by equiaxed
crystals. Microsegregation may sometimes be overcome by
annealing, but macrosegregation persists through subsequent
heating and working operations. So-called pipe segregates
occur around the pipe cavity. In normal segregation in steel,
the constituents (solute) in the iron (solvent) rejected from
the freezing liquid accumulate at the advancing solid/liquid
interface so that the constituents of lowest melting point
concentrate in the last portions to solidify, but in inverse
segregation this is reversed, for theliquid with high solute
concentration becomes trapped in between the dendrites thereby
causing a decrease in concentration of solutes from the ingot
surface toward the center. Inverse segregation, then, is a
concentration of constituents or impurities to a higher degree
near the outer surfaces (as compared to the interior) of an
ingot or casting.
Prior art methods of casting steel have provided cast
steel products having a relatively high degree of segregation
of impurities and alloying material~ within the cast steel.
Because of the high level of constituents and impurities in
steel, inverse segregation normally occurs. Such uneven dist-
ribution of impurities and/or constituents within the cast
steel makes it desirable that the total amount of same within
the steel be reduced so that subsequent processing of the cast
steel does not result in unacceptable internal and surface
characteristics in the product manufactured from the cast
steel. A reduction in the total amount of impurities, how-
ever, usually requires expensive additional refining of the
steel prior to casting and is sometimes commercially unfea-
sible or impractical altogether while sometimes the addition
of particular constituents (including alloying elements) is
desirable or necessary.
Among the impurities and alloying elements within the
X
,
~t79~73
-- 3 ~
steel likely to become segregated in prior art products are
sulphur, oxygen, phosphorus, manganese, silicon and carbon.
Any significant segregation o~ same will make it less commer-
cially useful. For example, significant segregation may cause
non-uniformity of tensile strength within the steel and make
it less suitable for subsequent drawing into wire. Segre-
gation of gaseous impurities may result in areas of porosity
near the top surface of the cast product, which, among other
drawbacks, causes inferior sheet surface quality. (See Whit-
more, B.C. and Hlinka, J.W., "Continuous Casting of Low-Carbon
Steel Slabs by the Haselett Strip-Casting Process", Open
Hearth Proceedings, 1969.) Severe microsegregation of manga-
.
nese will cause problems in many end products made from the
continuously cast billet due to its great effect on the auste-
nite to pearlite/bainite/martensite transformations. For
example continuously cast billets rolled into wire rod often
have high concentrations of manganese in the core which will
pron.ote the local formation of martensite during casting, thus
causing frequent breakage during the subsequent wire drawing
process.
This problcm has long been known in the art but there
have been few publications found by the inventors which dis-
close actual production data. A related study by Hans Van
Vuuren of the South African Iron and Steel Industrial Corpora-
tion, Ltd. tcopy contained in pages 306 to 334 of Steelmaking
Proceedings, Vol. 61, Chicago Meeting, April 16-20, 1978.)
illustrates one approach to controlling microsegregation and
its effects in a final rod product.
Van Vuuren concluded that central martensite could
usually be avoided in some steel wire rod by limiting the
total amount of manganese to 0.75% maximum, phosphorus to
0.020~ maximum, and then controlling the cooling in the
cooling line subsequent to the rolling process. There was no
mention of the microsegregation values of the continuous cast
blooms (ISCOR is believed to start by continuously casting a
205 mm x 315 ~m bloom on a Concast Bloom ~achine) but the
final rod product was analyzed and showed manganese microse-
`..~
~7947;~
-- 4 --
gregation values ranging from 101.5~ to 139.0% of the baseanalysis. Because of the extensive thermal diffusion between
the time of casting and the time of rolling, it is believed
that the microsegregation values of manganese in the cast
bloom wou~d have been much greater than in the final rod
product.
Up until now the prior art methods of solving problems
due to segregation (e.g., wire breakage) involved repairing
(e.g. homogenizing) the intermediate products (e.g., rod)
instead of avoiding the initial segregation during casting.
One reason for this is believed to be that it is much more
difficult to control the commercial high volume continuous
casting process.
In prior art ingot casting methods, segregation occurs
as the molten steel slowly solidifies, the impurities being
allowed to float by gravity separation to the top of the
ingot. In hiyher quality applications, a resulting concentra-
ted layer of impurities and/or solidification cavity at the
top of the ingot sometimes had to be physically removed
(scalped or scarfed) before the cast steel could be further
processed (see, for example, "recent Developments in Machine
Scarfing of Continuous Cast and Rolled Steel", Iron and Steel
Engineer, ~anuary, 1978, pgs. 68 71 and U.S. Patent No.
4,155,399, col. l,.lines 61-68.
Methods of continuous casting of steel have` been deve-
loped to avoid the handling of a large number of ingots and
the necessity of removing the top surface layer. In what the
applicants consider to be the most material and most commer-
cially accepted prior art method of continuous casting of
steel, molten metal is poured into an open ended vertically-
disposed mold constructed of a highly conductive material such
as copper, within which water is circulated for cooling pur-
poses. As the periphery of the metal solidifies to form a
shell of solidified steel adjacent to the mold wall, the
strand of steel is slowly withdrawn from the bottom of the
mold while molten metal is continuously poured into the top of
the mold. This type of process is sometimes reerred to as
~:L7~3
-- 5 --
the Junghans-type or Junghans-Rossi-type of continuous casting
system and has been successfully commercialized by Concast
A.G. of Zurich, Switzerland and Koppers Co., Inc. of the
United States, for example. An early Junghans patent is U.S.
Patent No. 2,135,183 (U.S. Class 164-83). Even here, however,
a surface may need to be scarfed for certain applications -
see U.S. Patent No. 4,155,399 (U.S. Class 164-82).
In the Junghans-type process the mold may be vertically
oscillated along a short path so that the mold moves with the
strand during each downward oscillation to increase the heat
transfer during the times when there is no relative movement
between the strand and the mold. Such oscillation increases
the possible speed of casting but often creates undesirable
oscillation marks or rings extending around the casting on the
surface thereof. `
As the strand leaves the mold, water sprays are normally
directed onto the surface of the semi-solid strand to complete
solidification thereof. In order to reduce the vertical
height requirement of the building containing the Junghans-
type casting machine, guide rollers have been utilized to bend
the strand through an arc of approximately 90 about a radius
of, for example, forty feet, and then to rebend the strand so
that it extends horizontally for cutting or subsequent proces-
sing. To avoid bending the strand twice in this fashion, and
to be able to install the caster in a smaller building, curved
molds have been developed so that the strand emerges from the
mold con~orming to the desired curved path and then is
straightened in one bending step to a horizontal orientation
for cutting.
A very readable exposition of traditional continuous
steel casting is provided in the December 1963 issue of
Scientific America magazine, "The Continuous Casting of
Steel," by L. V. Gallagher and B.S. Old, Vol. 209, ~o. 6, pp.
74-8~.
It will thus be seen that casting according to prior art
vertical mold processes does not rapidly (it usually takes a 5
story building or more) change the orientation of the solidi-
~ 7-
~L~7~qL73
fying steel, and allows molten metal in the center of the
strand to, at times, remain in a horizontal attitude as shown
in U.S. Patent No. 3,542,115 (U.S. Class 164-82) assigned to
Concast Incorporated, for example. Thus, impurities have an
opportunity to float upwardly during the progress of solidifi-
caction and in general, segregation occurs which sometimes may
be observed in a (long) transverse section of a prior art 4
inch x ~ inch square bar as a line of segregation occurring
about 1 inch in from the inner radius of the strand.
On a research basis, steel has also been continuously
cast in relati~ely horizontal molds, this being performed on
twin-belt casting machines similar in principle to the early
Hazelett Strip-Casting Corp. machine as shown in U.S. Patent
No. 2,640,235 (mentioned in the Whitmore and Hlinka publi-
cation). These two authors reasoned that because of the
influence of gravity and the approximately 20 from horizontal
attitude of the steel strand during solidification in these
research projects, the impurities within the steel tended to
rise and form a substantial zone of segregated material near
the top surface of the casting. These two authors state that
coupled with the top-surface oxide pit condition was an inter-
nal oxide segregation noted in macroetch tests of transverse
sections. Although the oxide segregation varied in degree,
the profile was similar in all slabs cast and the authors con-
cluded that from this data it was apparent that the oxide seg-
regation was severe enough to cause inferior sheet surface
quality. A number of possible solutions were tried, such as
concentrating on the elimination of mold-pool slag, use of
stationary, water-cooled copper edge-dams, use of submerged
feed tubes, etc., but the authors admitted that all these
attempts at a solution proved fruitless. It was reasoned that
heavy concentrations of segregating oxides were trapped in the
solidifying slab at the time when the top skin was between 1/2
to 3/4 inch thick. The authors concluded that casting at a
20~ angle resulted in a metallurgically unacceptable product
for sheet application produced from either Al-killed or
vacuum-treated steel because no way to remove these oxides was
~79473
found and the inclusions segregated toward the upper part of
the cast billet. They suggested going back to operating the
mold in the prior art (Junghans-type) vertical position as a
possibility for overcoming the segregation problem. It is
believed by applicants that the off-center and segregated dis-
tribution of the constituents and impurities also caused
unpredictable variations in subsequent attempts at processing
such as hot-rolling of the material.
SUMMARY OF THE INVENTION
Generally described, the present invention provides an
as-cast steel bar characterized by, when compared to the prior
art, a novel lack of segregation of manganese, oxygen, sulphur
and carbon. On the contrary, the present invention provides
an as-cast steel bar characterized by a particularly uniform
distribution of manganese, oxygen, sulphur and carbon. The
cast steel bar of the invention comprises steel that, when
viewed in (long) transverse cross-section for macrosegre-
gation, displays a maximum average variation in oxygen content
less than about 20 ppm (0.0020%) and an oxygen segregation
standard deviation less than about 8 ppm (0.0008~) on samples
containing about 0.01% oxygen, a maximum average variation
in sulphur content less than about 0.004% (40 ppm) and a
sulphur segregation standard deviation less than about 0~001%
(10 ppm) on a sample containing about 0.02% sulphur, and a
maximum average variation in carbon content less than about
0.01% (100 ppm) and a carbon segregation standard deviation of
less than about 0.004~ (40 ppm) on a sample containing about
0.185% carbon, and an improved as-cast tensile strength and
elongation. Similar good results are expected for Si, P. Cr
and other alloying elements normally used in steel. Microseg-
regation analysis has been carried out for C, Mn, S, Cr, and
Si by electron microprobe. The results indicate much less
microsegregation of manganese compared to prior art Concast
samples. For example, heats of molten steel, containng about
.46% carbon and about .98~ manganese, were cast both by the
method disclosed herein and by the wellknown prior art Concast
process. Samples of the as-cast bars were sectioned transver-
~7~73
sely and small specimens were cut from equivalent locations
about one-half the distance from the edge toward the center.
Thus neither the best nor the worst areas were selected for
comparison. The small specimens were mounted for electron
micro-probe examination and analyzed to determine the concent-
ration of manganese along a random line, about 1800 microns
long.
This procedure is generally well-known in the art and is
believed to be the easiest method for detecting microsegre-
gation in metals. Basically the process involves bombardingthe specimen with a small diameter beam of high energy elec-
trons which cause the specimen to give off characteristic
x-rays corresponding to the concentration of the elements
present. The x-rays are analyzed by diffracting them with a
crystal (to select one element at a time) then measuring their
intensity with any appropriate detector. Concentrations of a
specific element may be determined by comparing the relative
intensities of t~e x-rays generated by the element in the
specimen to a known standard. To obtain maximum accuracy,
about one-half of one percent, the ratio should be corrected
for absorption and flourescence of the emitted x-rays. Alter-
nately, the average intensity level can be assumed to repre-
sent the base concentration obtained by a normal chemical
analysis and then the variations in intensity directly repre-
sent variations from the base concentration.
When this latter procedure is used to compare a number
of specimens, it is useful to assign one value to each speci-
men which indicates the average intensity of the significant
variations and thus the average variation in concentration
from the base level. This value may be called the microsegre-
gation value and is axpressed as a percentage of the base con-
centration.
In our comparison the base analysis of our sample was
about 0.98 manganese. The results indicated that the prior
art as-cast steel bar has a manganese microsegregation value,
expressed as a percentage of base analysisl of about 300~
while the manganese microsegregation value of the as-cast bar
!,i
~ 7~3~73
g
produced by the method of our invention was less than about
175%.
One important feature of the invention is that our novel
product can be made on a known, prior art casting machine
using methods of operation within the experimental s~ills of
those having ordinary skill in our art. Our preferred method
of practicing our invention consists of forming a moving
arcuate mold by rotating a casting wheel, having a peripheral
groove, on its central axis and moving a band along its length
into contact with the peripheral groove at the upper part of
the casting wheel, moving the band and wheel in conjunction
about the lower portion of the wheel, and moving the band away
from the wheel (thusly forming one type of endless moving sur-
face-type mold), pouring molten steel into the arcuate mold,
cooling the mold to cause the molten steel to solidify in the
arcuate mold, withdrawing the cast bar from the arcuate mold,
normally additionally cooling the cast bar by the use of an
after-cooler after the cast bar has exited the closed portion
of the mold, and progressively stra:ightening the cast bar as
it moves away from the arcuate mold. See for example U.S.
Patent No. 3,623,535 (U.S. Class 164-87) and U.S. Patent No.
359,34~ tU-S. Class 164-263).
Thus, it is an object of the present invention to pro-
vide an improved as-cast steel product.
It is a further object of the invention to proviae a
novel continuous cast steel bar characterized by a lack, to an
unusual degree, of segregation or inverse segregation of
impurities or constituents within the bar.
It is a further object of the invention to provide a
cast steel bar more suitable than prior art bars for subse-
quent processing such as rolling into rod and drawing into
wire or hot-forming as by forging.
It is a further object of the invention to provide a
method for continuously casting a steel product having
improved internal properties.
Other objects, features and advantages of the present
invention will become apparent upon reading the following
.
~7~ 3
-- 10 --
specification when ta~en in conjunction with the accompanying
drawings.
BRIEF DESCRIPTION OF THE_DRAWINGS
Fig. 1 is a schematic diagram illustrating one example
of prior art apparatus suitable to produce the cast steel
product of the invention, the apparatus including a casting
machine having a rotatable casting wheel defined by a peri-
pheral groove therein and an endless band covering a portion
of the length of the groove so as to form a closed mold over
that portion.
Fig. 2 is a graph showing sulphur distribution in a
steel bar cast according to our invention.
Fig. 3 is a graph showing oxygen distribution in a steel
bar cast according to our invention.
Fig. 4 i9 a graph showing oxygen distribution in a steel
bar cast according to the prior art Hazelett twin-belt
process.
Fig. 5 is a graph showing oxygen distribution in a steel
bar cast according to another prior art process, the Junghans-
type process, and in particular, 2L Concast machine of the
commercially successful type.
Fig. 6 is a graph showing carbon distribution in a steel
bar cast according to our invention.
Fig. 7 is a histograph comparing the tensile strength o~
the novel cast steel product of our invention to another cast
steel product produced by a prior art process.
Fig. 8 is a graph showing the x-ray intensity variation
due to microsegregation of manganese in a steel bar cast
according to our invention.
Fig. 9 is a graph showing the x-ray intensity variation
due to microsegregation o~ manganese in a steel bar cast
according to the prior art Concast process.
DESCRIPTION OF THE PREFERRED EMBODIMENT
Referring now in more detail to the drawing, in which
like numerals of reference illustrate like parts throughout
the several views, Fig. 1 shows a casting wheel apparatus 10
for producing product of the present invention. This appara-
~7~7~
tus is similar to that disclosed in U.S. Patent No. 3,623,535
(U.S. Class 164-87) for example. A casting wheel 10 defines a
peripheral groove G therein which is covered, for a portion of
the periphery of the casting wheel, by an endless flexible
band or belt 11 to form a closed mold M. The 1exible band 11
is held against a portion of the periphery of the casting
wheel by band support wheels 12, 13 and 14 and moves with the
wheel 10 as it rotates. Near the band support wheel 12 where
the closed mold M begins, molten steel is discharged from a
pouring pot or tundish 16 into the mold M through a spout
16a. In our preferred embodiment, all exterior surfaces of
the casting wheel and band are continuously cooled by a spray
of coolant fluid, the outer portion of the groove and band
being cooled by cooling sprays from nozzles (not shown) from
headers or manifolds S2, S3, S4, and the inner portion of the
peripheral groove G being cooled by sprays from the nozzles on
header Sl. The spray of each nozzle (or groups of nozzles)
along the inner side of the peripheral groove may be indivi-
dually adjusted to vary the volume of cooling fluid sprayed
therefrom, and thus vary the rate at which the metal within
the mold M is cooled. A supply of coolant fluid to the
nozzles (or groups of nozzles) may likewise be controlled by
adjustable valves to allow starting and stopping of the
coolant flow and to permit variation in the total volume of
coolant flow. See, for example, the cooling arrangement
suggested by Southwire Company's U.S. Patent ~o. 3,279,000
(U.S. Class 164-433).
An extended bending section 18 is positioned beyond and
above the band support wheel 14. The bending section 18
~0 serves as a means for straightening the cast steel bar B with-
drawn from the peripheral groove of the casting wheel 10 after
exiting the closed portion of mold M. The bending section 18
includes a plurality of support guide rolls 19 mounted on a
frame (not shown). Side guide rolls (not shown) may also be
utilized in the bending section 18 to confine the cast steel
bar approximately to a vertical plane. Although the support
guide rolls 19 ~ay be either driven or non-driven, we find it
` f
~L~7~473
- 12 -
is preferable that at leas-t some of the support rolls 19 be
driven to assist in the straightening of the cast bar.
In our preferred embodiment an after-cooling header 21
is located above and adjacent the band support wheel 14 to
apply a direct spray of coolant fluid onto the cast steel bar
emerging from the arcuatP mold M.
In the operation of the system according to the method
of the invention, the casting wheel 10 is rotated in a
counter-clockwise direction and molten steel is poured from
the tundish 16 through the spout 16a into the closed mold M
formed between the peripheral groove G of the casting wheel
and the flexible band 11. Molten steel is poured in a cont-
rolled manner well known in the ferrous and non~ferrous
casting arts into the mold M at a rate so that the rotation of
the casting wheel moves the steel in the mold M away from the
spout 16a as fast as the molten steel flows through the spout
to maintain the surface of the pool of molten steel at a cons-
tant level at the entrance of the mold M. The exit end of the
spout 16a is located as closely as possible to the entrance to
the mold M to allow the molten steel to flow directly from the
spout into the pool of molten metal in the mold.
As the molten steel is carried around the casting wheel
10 within the mold ~, coolant fluid is directed against the
mold from the no~æles in header S]. and the nozzles of the
other headers S2, S3 and S4, and the amount of coolant applied
to the band and casting wheel is adjusted as desired to cont-
rol the rate of cooling of the molten metal. Our preferred
embodiment has a condition of very uniform cooling about and
along the longitudinal axis of the cast bar, see for example
Southwire Company's U.S. Patent No. 3,279,000 to Cofer (U.S.
Class 164-433). Initial rapid cooling and solidification of
the molten metal occurs at the surfaces of the casting wheel
and band, causing the formation of a skin or shell of solidi
fied steel having an equiaxed grain structure. Continued
extraction of heat from the partially solidifie~ bar then
causes solidification of the metal within the molten core in a
progressive and uniform (including uniform at each pOiIlt
~79~73
- 13 -
around the periphery) manner to form a dendritic or
substantially e~uiaxed structure, depending upon the superheat
of the steel, from the shell toward the center of a solid
steel bar B.
The steel which entered the mold M as molten metal at an
upper portion of the casting wheel moves in a downward direc-
tion about the lower portion of the casting wheel and then in
an upward direction until it leaves the closed portion of the
mold M near the band support wheel 14, passes through the
after-coolant spray from the nozzles associated with the
header 21 and reaches a guide wheel 15, whereupon it is guided
away from the casting wheel. In our preferred embodiment, the
temperature of the exterior surface of the peripheral skin of
the solidified steel as it emerges from the closed portion of
the mold M does not exceed about 2500F. but is not less than
about 1900 or 2000F. The cast bar leaving the casting wheel
has a shape conforming to the curvature of the arcuate mold M
and therefore is progressively straightened by progressively
increasing the radius of the bar B as the bar moves through
the extended bending section 18. The guide rolls 19 support
the bar and guide it through its unbending or straightening
path above the casting wheel 10, at least one pair of the
guide wheels 19 preferably being driven to pull the bar B
along its length from the casting wheel 10. The molten core
of the bar B is completely solidified by the time it passes at
least the last coolant spray from the last nozzle on header 21
to assure that the bar is completely solid before reaching a
point which is on a level with ~he level of the pool of molten
metal at the entrance to the mold M. Thusly, the molten metal
in the core of the bar will not flow opposite to mold movement
through the unsolidified bar center thereby creating a void in
the center of the bar. The bar is thereby also solid and
sound metallurgically before entering the bending section 18,
and its temperature just prior to bending may also be adjusted
by adjusting the volume of coolant supplied by the header 21
in order to control the internal stresses of the bar during
straightening.
73
- 14 -
In the embodiment of the castin~ wheel 10 disclosed
herein, the mold M is approximately trape7oidal in shape with
small dimension located at the inner portion of the peripheral
groove and a large dimension located adjacent the band 11.
Thus the steel bar cast by a typical casting machine 10 may be
approximately 2-5/8 inches wide at its largest width, 2-1/8
inches wide at its smaller width and 1-7/8 inches deep, with
an approximately 1/4 inch radius joining the smaller width
with the two sides of the bar. Other bar sizes and shapes may
be cast as desired. To date, for example, applicants have
been successfuL in casting an approximately 4.8 square inch
bar at a speed of appriximately 44 feet per minute (528 inches
per minute) and an appriximate 8.1 square inch bar at a speed
of appriximately 35 feet per minute (420 inches per minute).
Applicants believe that the novel cast steel bar of the
present invention may be produced at a relatively high linear
speed because the relatively long length of the arcuate mold M
cooled by quick chilling coolant sprays allows solidification
to be achieved in spite of the relatively high rotational
velocities of the casting wheel 10. Furthermore, the rela-
tively small radius of the casting wheel 10 causes the orien-
tation of the molten steel to change rapidly as the wheel
rotates, in contrast to prior art commercially proven steel
continuous casting techniques wherein the solidifying steel
remains in a horizontal or appriximately horizontal orien-
tation for a substantial period, allowing impurities to ~loat
upwardly during the process of solidification. Applicants
believe that when casting a bar B with a relatively small
cross-sectional area according to the present invention, the
bar B may be quickly solidified by the coolant spray from the
nozzles associated with headers Sl, S2, S3, S4 and 21 before
any substantial segregation of constituents or impurities may
occur. Thus, the method of the invention whereby a relatively
rapidly rotating casting wheel having a relatively small
radius is cooled sufficiently to quick freeze impurities
before segregation and/or inverse segregation can occur,
produces the novel cast steel product of our invention, having
'
~7~473
properties significantly different from cast steel produced in
prior art continuous casting machines.
Applicants believe that the design of the wheel mold,
cooling water spray zones, and the relatively smaller cross-
section of the cast bar makes it possible to achieve a higher
rate of heat transfer compared to the prior art continuous
steel casting methods.
Some idea of the high rate of cooling or solidification
can be obtained from the metallurgical heights of the casting
systems. The metallurgical height is defined as the distance
between the top of the liquid pool in the mold to the point of
complete solidification. In our method, we have worked with a
metallurgical height of about 15 feet or less when manufac-
turing a 4.8 sq. in. bar at 35 to 44 FPM (420 to 528 I~M) and
for 8.1 sq. in. bar at 25 to 35 FPM (300-420 IPM). In the
conventional continuous casting systems of the Junghans-type
the metallurgical heights are generally reported to be about
50 to 70 feet for cas-ting of 4" by 4" billets at a speed of
100 to 120 inches per minute (8.33 to lO.0 FPM). We have
found that our cast bar becomes completely solid in about 25
to 30 seconds whereas we understand it requires about 6
minutes for complete solidification of the Junghans-type bar.
We believe that our fast rate of cooling reduces the
flow of liquid of higher solute concentration into the
inter-dendritic channels and thereby reduces the inverse
segregation while the non-metallic impurities present in the
liquid steel fre~ze with the liquid with a random
distribution.
We also believe that the ~uick orientation change of
molten steel in the moving wheel mold reduces the chances of
segregation of impurities at an undesired position within the
cast bar. That is, offering an example, as the cast bar with
its initial relatively thin frozen shell and large molten
center moves counter-clockwise from opposite approximately
manifold S2 (see Fig. 1), past manifolds S3, S4, and then 21
-to a point where the frozen shell has become quite large with
respect to the molten center, which means that the cast bar
!~
~7~473
- 16 -
has moved through an arc of more than 90 and preferably more
than 180, applicants believe that such orientation change of
the casting throughout the course of solidification tends to
eliminate the formation of heavy concentrations of segregated
constituents or impurities which would otherwise normally
float to the upper portion of the intexior of the solidifying
shell, because, essentially, the "upper portion" of the soli-
difying shell is always changing throughout the rotation of
the wheel.
Measurements have been made to determine the degree of
segregation of sulphur and oxygen impurities, and the degree
of segregation of carbon, in novel cast steel bars that were
cast according to the method disclosed. To establish a segre-
gation pro~ile from the wheel side of the cast bars to the
band side of the cast bars, three sets of samples for analysis
were punched from (long) transverse sections of the as-cast
bar, one at the center of the section, one 20 mm to the left
of the center and one 20mm to the right of the center. The
values obtained were then averaged to obtain an average pro-
file for each bar. Such results are shown in Figs. 2, 3 and
6. Applicants believe that those in the art use "(long) tran-
sverse" and just "transverse" interchangeably if there is no
likelihood of confusion with a short transverse section (see
ASTM Designation E399-74, Crack Plane Orientation Identifi-
cation Code, for example).
Fig. 2 is a graph of average percentage sulphur composi-
tion vs. position between the wheel side of the bar (Omm) and
the band side of the bar (44mm in this case), the steel having
a composition by weight of approximately 0.45% carbon, 0.02%
sulphur, 0.99% manganese, 0.02% phosphorus and 0.21% silicon
(Specimen $45). The maximum variation in average sulphur con-
tent across the bar was 0.0013% (13 ppm), for the measurements
shown in Fig. 2, and the standard deviation was 0.000498%.
Applicants believe this represents an unexpectedly high
uni~orm distribution of ~ulphur with no significant delete-
rious segregation. Tests of other specimens of the novel
product indicated maximum average sulphur content variations
,. ~
.
~7~4~3
- 17 -
ranging from 0.00114% (11.4 ppm to 40 pmm), and sulphur
standard deviations varying from 0.000483% to 0.00138% in
samples having 0.01755% and 0.02993% sulphur, respectively, as
shown below.
Average of 3 sulphux measurements
at each position listed in PPM
Specimen # = 26 41 43 45 48
5mm = 170.31308.0234.3 226.7 316
lSmm - = 18~.7310.6233.0 240.0 328
1025mm = 174.3271.0223.7 235.0 316
35mm = 170.7297.6227.0 239.7 317
40mm = 181.7297.6231.0 238.7 325
47mm = -- 311.0 238.7 -- --
Average = 175.5299.3231.2 236.0 320.4
Range = 11.440.0 11.7 13.0 12.0
Std.
Deviation = 4.8313.8 4.88 4.98 5.03
Fig. 3 is a graph of average oxygen content (in pmm)
vs. position between wheel and band sides for the same cast
bar, #45, measured in Fig. 2. The oxygen content was
approximateiy 70 ppm ~0.007%) and the maximum variation in
average oxygen content across the bar, as shown in Fig. 3, was
5 ppm (O.OaO5~), and the standard deviation was 1.651 ppm.
This again represents an unexpectedly good result - a very
high degree of uniformity of constituents in the structure of
the novel cast steel bar. It should also be noted that center
porosity, sometimes present in a continuously cast steel bar
(even ours), may contribute to the measured oxygen content at
that particular location in the bar. Applicants believe that
such porosity does not represent true segregation to those in
the art and is generally healed during subsequent hot proces-
sing.
`~
~L~7~73
- 18 -
The improved oxygen segregation properties of the
present invention may be seen by comparing Fig. 3 to Figs. 4
and 5. Fig. 4 is a graph of percentage of oxygen content vs.
position between the bottom and the top of a cast bar cast
using a Hazelett Strip-Casting machine having a substantially
horizontal mold. The graph is ta~en from page 43, Fig. 6 of
Whitmore, B.C. and Hlinka, J.W., "Continuous Casting of Low-
carbon steel Slabs by the Hazelett Strip-Casting Process,"
Open Hearth Proceedings, 1969. Converting to a ppm basis,
_
Fig. 4 shows a maximum variation of approximately 100 ppm
(0.01%) or more and a standard deviation of approximately
29.88 ppm. This approximately 20 from horizontal Haselett
experimental mold process thus produced a cast bar with a
significant oxygen segregation problem even when the average
oxygen content was relatively low, about 0.004%. The worst
segregation was located near the top surface of the bar as is
evident from Fig.-4 herein and the Whitmore and Hlinka publi-
cation.
Fig. 5 is a yraph of oxygen content vs. position for a
cast steel bar produced by a Concast vertical mold continuous
casting machine including an arcuate oscillating mold. The
graph represents an average of five samplings taken in (long)
transverse section from the bottom to the top of the bar,
which had a composition by weight of 0.46% carbon, 0.94% man-
ganese, 0.021% phosphorus, 00016~ sulphur and 0.22% silicon.
The maximum variation shown in Fig. 5 is approximately 26.5
ppm, and the standard deviation is 10.6 ppm. The average oxy-
gen content is about 0.006%. Another specimen, having an
average oxygen content of about 0.009%, showed a variation of
29 ppm.
A cast bar produced according to the method of the
invention also has an unexpectedly uniform carbon distribu-
tion, as shown in Fig. 6 which is a graph showing an average
profile of the carbon content of such a cast bar (Specimen
#48). This particular bar had a composition by weight of
approximately 0.185% carbon, 0.59~ manganese, 0.01% phos-
phorus, 0.032% sulphur, and 0.17% silicon. The points plotted
~7~73
-- 19 -- .
in Fig. 6 are averages of three measurements for each position
across the cast bar between wheel and band sides, as was the
case for Figs. 2 and 3. Fig. 6 shows a maximum average carbon
content variation across the bar of approximately 0.009~ (90
ppm), and a standard deviation of 0.00305%. In accordance
with this invention, the steel melt is preferably prepared
from a chemical system which has carbon present as one of the
constituents in a range of approximately 0.04~ by weight to
1.4~ by weight. The maximum variation for samples in this
range at variance with our specimens are expected to be pro-
portional. Applicants have ~ound that particularly superior
results are achieved when the carbon content of the steel is
between about 0.06% and 0.80% by wei~ht.
Still further measurements have been made to determine
the as-cast tensile strength of the novel cast steel bar of
the invention, and to compare it to the as-cast tensile
strength of a steel bar cast using a prior art commerciall~
proven Concast machine including an oscillating, arcuate mold,
both steel bars having been cast from the same steel melt.
The measurements were carried out at a strain rate of 0.001/
second using an one inch Extensometer. Fig. 7 is a histograph
showing the tensile strength of a sample of the novel cast bar
to be approximately 107-110 ksi (107,000-110,000 pounds per
square inch) as compared to that of lthe prior art Concast bar
which had a tensiie strength of approximately 93-94 ksi. The
composition of the steel on this melt was 0.45% carbon, 0.97
manganese, 0.019% phosphorus, 0.017~ sulphur, and 0.21~ sili-
con. Applicants believe that the approximate 10%-15% or more
increase in tensile strength of their novel product is consis-
tent with the unusually uniform distribution of constituentsand impurities observed in the novel product, as described
above. In these same tests it was also observed that the
novel as-cast bar had a greater percent elongation and a
greater proportional limit in ksi than the prior art Concast
bar, see below:
Proportional
Test Sample Tensile Strength ~ Elongation Limit
Novel 1 107 Xsi 10 64.3 ksi
Bar2 110 ksi - 13 71.7 ksi
3 108 ksi 16 72.0 ksi
, .
~ 20 _ ~7~'~73
Prior 4 94 ksi 8 62.0 ksi
Art Bar 5 93 ksi 8 57.8 ksi
Fig. 8 graphically shows the intensity of characteristic
manganese x-rays, along a random line of about 1800 microns in
length, within a specimen of steel bar cast by the present
invention. It can be seen that the intensity level is rela-
tively constant about the line marked 100~ which corresponded
to the base concentration of 0.98% manganese and which also
corresponded to an absolute reading of approximately 11 units
on the graphical readout from the electron microprobe
analyzer. Th~re is only one significant variation which
measures about 173% of the base leve; i.e., equivalent to a
local concentration of about 1.69% manganese.
Fig. 9 graphically shows the manganese x-ray intensity
in a specimen cast by the prior art Concast process. It
should be noted that the intensity level contains many peaks
which correspond to small area~ of segregated manganese.
Here, an absolute reading of approximately 12 units was
observed on the strip chart readout from the electron micro-
probe unit and used for the 100% li~e. The average value ofthe most significant peaks is about 320% of the base level and
the maximum variation in manganese content observed was over
400%, see below:
Peak % of Base Level
#1 396%
2 227~
3 292%
4 404%
262%
6 335%
Applicants have found that particularly superior results
are achieved when the manganese content of the steel is
between about 0.30% and 1.20~ by weight.
While this invention has been described in detail with
particular reference to preferred embodiments thereof, it will
be understood that variations and modifications can be effec-
ted within the spirit and scope of the invention as described
- 21 - ~ ~7~473
hereinbefore and as defined in the appended claims. For
example, applicants have reported herein only a representative
sampling of the infinite numher of steel compositions which
may be cast according to the present invention. For other
compositions in which the concentrations of the constituents
and impurities are at variance with the exact concentrations
in the specimens analyzed, applicants expect proportionally
improved results~
