US 3863702 A
A method of centrifugally casting hollow metal objects is provided in which molten self-disintegrating slag is added to the rotating mold immediately after casting molten metal and transform in the solid state giving a volume change which causes the slag to disintegrate therein.
Description (OCR text may contain errors)
United States Patent Hallerberg et al. Feb. 4, 1975  CENTRIFUGAL CASTING METHOD 3,293,708 12/1966 Fruitman 164/114  Inventors: William L. Hallerberg; Gilbert W. 2 5322 85 Gordon, both of Kokomo, Ind.
 Assignee: Cabot Corporation, Kokomo, Ind. rim ry Ex miner-Andrew R. Juhasz Assistant Examiner-Rising  Flled' 1973 Attorney, Agent, or Firm-Buell, Blenko and  Appl. No.: 323,104 Zie enheim  U.S. c1. 164/114, 164/55, 148/30,  ABSTRAPT 143 27 143/19 A method of centnfugally casting hollow metal ob ects  Int. Cl. B22d 13/00 is Provided in which olten Self-disintegrating slag is-  Field of Search 164/114, 118, 55 added to the rotating mold immediately after asting molten metal and transform in the solid state giving a  Refer Cit d volume change which causes the slag to disintegrate UNITED STATES PATENTS 1,831,310 11/1931 Lindemuth 164/114 XR 7 Claims, 4 Drawing Figures PATENTEU E 4W5 3.863.702
SHEET 2 BF 2 l CENTRIFUGAL CASTING-METHOD This invention relates to centrifugal casting methods and particularly to a method of centrifugal casting in which a self-disintegrating slag is introduced into the centrifugal mold immediately following the molten metal to form an inner lining in the cast metal. By the term self-disintegrating slag we mean a slag which on solid state transformation gives a volume change which causes the slag to disintegrate or break up.
Centrifugal casting of metal into a variety of generally concentric hollow articles such as pipe is well known and has beenpracticed for many years. It is used in casting gray iron pipe as well as pipe from various types of steels and superalloys depending upon the ultimate use to which the article is to be placed. In the case of steel and superalloy pipe and other hollow articles made of steel or superalloy by centrifugal casting techniques there has long existed a problem caused by shrinkage at the internal diameter of the pipe resulting in porosity and poor internal wall surface smoothness. Such pipe had a high percentage of scrap because of leakage and because of poor surface finish. No satisfactory solution to this problem existed except to cast the pipe to a thicker wall and smaller internal diameter than desired and then to machine out the defective inner wall surfaces down to solid metal to eliminate internal diameter porosity and to provide a good internal surface finish. This is obviously costly both in machine time and labor and in scrap losses.
We have devised a method of overcoming this problem and producing centrifugally cast articles substantially free from internal porosity and with good internal surface. The system is adaptable to any type of centrifugal casting.
We provide a method of centrifugal casting which comprises the steps of introducing molten metal into a rotating centrifugal mold, immediately introducing a molten self-disintegrating slag into the rotating centrifugal mold after the molten metal to form an inner slag lining in the centrifugally cast metal, cooling to cause the metal to solidify and the slag to solidify and disintegrate and removing the slag from the cast metal article. Preferably the slag is introduced into the mold immediately after metal pouring ends, and preferably within five seconds after the metal pour ends. The slag is preferably above about 3,000F and the slag-to-metal weight ratio is about 0.3 or more. We prefer to use a slag of di-calcium orthosilicate (Ca SiO diluted with calcium fluoride to reduce its melting point. Such a slag will decrepitate upon cooling and can be readily removed e.g. poured from the finished cast pipes. Other self-disintegrating slag can be used, if desired; however, we have found di-calcium orthosilicate to be the preferred slag.
in the foregoing general description of our invention we have set out certain objects, purposes and advantages of our invention. Other objects, purposes and advantages will be apparent from a consideration of the following examples and the accompanying drawings which are hereafter described.
EXAMPLE I metal. *HAYNES is a registered trademark of Cabot Corporation.
The alloy was poured at 2,850F into the mold turning at rpm. At the end of the metal pour, the mold speed was increased to 1,150 rpm and then reduced to '900 rpm. One hundred and forty pounds of alloy were used per tube. For the slag tubes, 35 pounds of molten slag were used. Results of the experiment including the time delay between the introduction of the metal and the introduction of the slag, the slag temperature in the furnace, the wall thickness measurements and the depth of shrinkage determined metallographically are shown on Table l. Pipes N0. 1, 4 and 9 were made without slag and should be considered standard pipe for this size. The time of slag introduction was measured from the end of the metal pour. The slag temperature was measured in the furnace by optical pyrometer. Wall thickness measurements were made with calipers approximately 3 inches from the end of the tube. Depth of shrinkage was measured metallographically by polishing a cross section of the tube and examining it under the microscope. The numbers indicate the maximum depth of shrinkage found.
TABLE I RESULTS OF HAY-NES" HL-40 ALLOY, HEAT 2351 SLAG CENTRIFUGAL EXAMPLE 1 Metallographic Time of Slag Furnace Slag Wall Thickness Depth of Shrinkage Pipe lntroduction, Temperature Hot End, Cold End. Hot End, Cold End, No. Seconds F lnch Inch lnch lnch 1-1 No slag added See Table 11 0.080 0.070 1-2 4.0 2950 /a A 1-3 5.0 3050 See Table ll 0.040 0.055 1-4 No slag added 1-5 4.5 3050 1f; 9/32 1-6 6.5 3050 5/16 1-7 8.0 3 I00 as /1; 1-8 8.0 3100 See Table II 0.075 0.085 1-9 No slag added Vs 1-10 4.5 3150 9/32 '"End of metal pour was F0 'Mcasurcd optically in furnace After cutting off an air test specimen, pipes No. l, 3 and 8 were sliced lengthwise and the wall thickness was measured with micrometers at one foot intervals along the pipe. The measurements are given in Table 11. The acceptable range of wall thickness for this size tube is three-eighths-inch +one-sixteenth or 0.375 to 0.4375 inch.
TABLE 11 WALL THICKNESS MEASUREMENTS OF 454-1NCH OD HAYNES: HL-40 ALLOY CENTRIFUGALLY CAST PIPE 'Cust without slag 2. Cast two pipes from each heat, 4 inches O.D. by 0.400-inch wall by 8 feet in length without slag as a standard.
3. Cast l6 pipes, each, 4 %-inches O.D. by 0.400-
inch wall by 8 feet in length, using the instant slag centrifugal invention.
All pipes had 12 inches cut from the cold end (opposite the pouring spout) and these sections cut in half, with the half away from the cold end sent for air test evaluation. The cold end of each pipe had a one inch thick ring cut from the end away from the cold end. These rings were sectioned into four equal segments and submitted for metallographic examination to determine the shrinkage porosity.
After all pipes were sectioned for the air test evaluation and the metallographic examination, several pipes were sectioned lengthwise for wall thickness checks. Pipe Nos. 2-1, 2-2, 2-4, 2-6, 2-7 and 2-10, from Heat No. 2,551, were sliced lengthwise and the wall thickness was measured with micrometers at the ends and at one foot intervals along the pipe. Table 111 shows the measurements along the pipe.
TABLE 111 WALL THICKNESS MEASUREMENTS OF 4' /4-1NCH O.D. HAYNES HL- ALLOY CENTRIFUGALLY CAST PlPE FROM HEAT 2551 Wall Thickness (inch) Position from Pipe Pipe Pipe Pipe Pipe Pipe Cold End, Feet No. 2-1" No. 2-2** No. 2-4 No. 2-6 No. 2-7 No. 2-10 l (Cold End) 0.239 0.406 0.398 0.389 0.365 0.380 2 0.486 0.405 0.392 0.394 0.388 0.387 3 0.565 0.407 0.397 0.402 0.390 0.391 4 0.572 0.418 0.409 0.409 0.406 0.405 5 0.563 0.412 0.404 0.409 0.415 0.410 6 0.348 0.402 0.409 0.41 1 0.425 0.416 7 0.253 0.402 0.410 0.405 0.434 0.415 8 (Hot End) 0224 0.386 0.409 0.391 0.434 0.415
Acceptable wall thickness range 0.375 to 0.4375 inch Cast without sla minted at 900 rpm. Cast without slug. rotated at 1100 rpm.
EXAMPLE 11 Two 1,500-pound heats of HAYNES HL-40 alloy were cast into twenty-one tubes, 4 A-inches CD. by %-inch wall thickness by 8-feet long for this experiment. Sixteen pipes were cast using the slag centrifugal technique of adding molten slag after metal pour, and five pipes were cast without the slag addition as a standard.
The normal practice for the centrifugal process is to pour the metal at 2,850F into a steel mold turning at rp'm.- After the metal pour, the mold speed was increased to 1,100 rpm and then reduced to 900 rpm until solidification. The first pipe cast in this series was poured per the standard practice, with the exception of the mold speed which was held at 900 rpm. It was found that 900 rpm is too slow because most of the metal stayed in the center of the pipe. The remainder of the pipes were cast following the standard practice with the exception that mold speed was maintained at 1,100 rpm until solidification had taken place.
Slag centrifugal casting procedures used for the series were as follows:
1. Melt self-disintegrating Slag and hold at 3200F or hotter.
The results of the air test can be seen in Table IV. The test pieces were machined to a selected wall thickness and air tested after each wall reduction. The test consisted of sealing each end of the pipe and submerging it in water. Compressed air is introduced at one end of the pipe and is maintained for a specified length of time. The air testing was performed according to Union Carbide Specification, CFTM-lOO, with the exception of the surface area tested. The specification calls for an inch minimum width, flatbottomed groove to be machined to wall thickness for testing. A three-inch wide. flat-bottomed groove was machined to wall thickness for testing in these experiments because it was believed that the three-inch wide groove would be a more severe test than the one-inch groove due to the greater surface area tested.
The taper of the pipes which were not sectioned lengthwise was obtained by taking five measurements (with micrometers) around the diameter of both the hot and cold ends of the pipe. These measurements were averaged and the difference between the two averages was considered the taper of the pipe and is shown in Table V as wall taper.
TABLEIV AIR TEST RESULTS (COLD END) 4%-INCH DIAMETER SLAG CENTRIFUGAL CAST PIPE Heat Pipe Wall Thickness Range No. No. and Failure, (Inch) 2-12* 0130-0140 2 l3 =5 3 2 |4 4 K t 2-l5 0015-0020 2-16 0015-0020 2-17 0.0l50.020 248* 0.|30 0.t40 2-19 0015-0020 2-20 0.015-020 '(nst without slug and should he considered sumdurd production pipe for this size.
'Pussed 0.020 inch wall thickness; unable to machine further.
The pipe samples submitted for metallographic examination were polished on the fine alumina wheels and then electropolished in a solution of 85 percent methanol and percent sulfuric acid. The results were obtained by measuring at lOOX magnification using a table with micrometer type traverse. The numbers indicate the maximum depth of porosity for each sample. The results of the metallographic examination in measuring porosity depth as determined by measurements taken on the polished pipe sections are shown in Table V.
TABLE V addition. FIG. 1 is a photograph of the polished and etched cross section of the cast pipe showing the condition as indicated by A in the figure. FIG. 2 is a photomacrograph at 8X magnification showing the carbon enriched condition present at the ID. of the pipe cast using the molten slag as indicated by B in the figure. H6. 3 is a photomacrograph at 8X magnification showing the shrinkage at the ID. of the pipe without molten slag as indicated by C in the figure. The amount of shrinkage in pipe cast without molten slag is noticeably greater than that with slag.
Hardness values were obtained on the cross section of several of the cast pipes and disclosed a difference in hardness values of approximately ten points, using the Rockwell A scale, between 1D. and OD. areas. Due to the amount of hardness difference, samples were submitted for a chemical analysis of the carbon at the [.D. and OD. Millings were machined from both locations (LD. and CD.) of the tube and the results were as follows: l.D. Carbon 1.20%; CD. Carbon 0.41%. Recheck ofLD. and OD. milling from another tube produced approximately the same results: 1.D. Carbon 1.10%, 0D. Carbon =0.44%. Carbon specification range for HAYNES HL-40 alloy is 0.35/045; consequently the pipe l.D. experienced a carbon enrichment. Chemical analysis of the heat revealed that all elements, even the carbon, were within the specified range for HAYNES HL-40 alloy indicating that the high carbon was probably picked up from the slag. It is thus evident that, due to the high temperature and the length of time the molten slag was held in the graphite furnace liner, the slag had picked up carbon from the liner and carbon exchange had taken place between the molten slag and the metal at the ID. of the pipe.
In order simultaneously to achieve minimum air test shrinkage and minimum pipe wall taper, the results demonstrate the slag temperature must be raised to RESULTS OF HAYNES HL-40 ALLOY SLAG CENTRIFUGAL EXAMPLE 11 Time of Slag Wall Metallograhic Heat Pipe Introduction. Ladle Temp! Slag-Metal Taper Depth of Shrink No. No. Seconds of Slag, F. Ratio 1nch*** Cold End, lnch 2551 2-1 No slag added 0 0.080 2-2 No slag added 0.021 0.120 2-3 8.0 3000 0.2 0.030 0.020 2-4 8.0 2920 0.2 0.018 0.040 2-5 4.0 2800 0.2 0.007 0.030 2-6 4.0 2820 0.2 0.022 0.040 2-7 8.0 2820 0.3 0.069 0.025 2-8 8.0 2820 0.3 0.005 0.025 2.9 4.0 3000 0.3 0.011 0.035 2-10 4.5 3000 0.3 0.036 0.035
2552 2-11 No slag added 0 0.120 2-12 No slag added 0.008 0.130 2-13 8.0 3000 0.3 0.1 15 0.040 2-14 8.0 3000 0.3 0.098 0.030 2-l5 8.5 2900 0.2 0.088 0.040 2-16 8.5 2800 0.2 0.020 0.030 2-17 5.0 3000 0.2 0.030 0.025 2-18 No slag added 0.012 0.100 2-19 4.5 3000 0.2 0.042 0.030 2-20 4. 2820 0.3 0.027 0.040
Poured at 900 rpm. all others poured at 1100 rpm. "Difference between hot end and cold end wall thickness of B-t'oot tube. Slag timc is after end of metal pour. Tcmpcraturc is measured by immersion thermocouple in ladle prior to pour.
Further metallographic studies revealed a dark band around the [.D. of the pipe cast using the molten slag 3,000F or higher, the slag/metal ratio raised to 0.3 or higher and the slag time cut back to about four seconds mu of less, and all factors held at these levels with minim process .variation. In addition the main effect" heat differences in alloy HL-40 is seen to significantly affect the overall level of pipe taper at the 95 percent confidence level.
S fou urprisingly, no main effects or interactions were nd to be important in controlling metallographic the test method employed. The tubes are crosssectioned so that one may, or may not, cut into a particular section where porosity is a problem.
EXAMPLE lll Ten pipes were cast eight with molten slag of this invention and two without slag. The practice used was:
. Melt decrepitating slag to 3,1003,150F maximum. (Although the previous example showed high slag temperatures to be best for taper and shrinkage, temperature was reduced to reduce carbon pickup from the crucible.)
. Cast two tubes, 4 Winches CD. by 0.400-inch wall by 8 feet in length, without slag. Mold speed was 50 rpm, for trough, increased to 1,100 rpm until solidification.
. Cast eight tubes, 4 A-inches CD. by 0.400-inch by 8 feet in length using the instant slag centrifugal invention.
. Preheat all pouring ladies and tundish as hot as possible. The conditions and operations of the pipe during casting are set out in Table VI.
TABLE VI Specification wall thickness range 0.375 inch to 0.4375 inch Metallographic examination of the electropolished pipe cross sections revealed shrinkage depths of 0.035 inch to 0.100 inch. These results were obtained by measuring at 100X magnification using a table with micrometer type traverse and are listed in Table Vlll. Also no heavy line was present at the ID. of any pipe cast in Example 111. FIG. 4 is a photomacrograph at 8X magnification showing the absence of the heavy line found in FIG. 2.
TABLE Vlll 4%-|NCH DIAMETER SLAG CENTRlFUGAL CAST PlPE METALLOGRAPHIC SHRlNKAGE DEPTH MEASUREMENTS HEAT 2681 ULIBUJLAJWWWUJUW Poured with out slag Hardness values were taken on several pipe cross sections, from the 1D. to the OD. and disclosed a hard- RESULTS OF HAYNES HL- ALLOY HEAT NO. 2681, SLAG CENTRlFUGAL EXAMPLE 111 Air test, Time of Slag Ladle Temp. Slag Metal wall thickness Pipe Introduction, of Slag Metal Temp. range at failure, No Seconds "F Ratio "F inch 3-1' 2880 .l20.139 3-2* 2910 .l25.l41 3-3 5.1 2820 .3 2850 026-038 3-4 3.2 2780 .3 2880 .023-.034 3-5 12.6 2700 .3 2900 025-035 3-6 5.9 2850 .3 2900 030-038 3-7 5.5 2800 .3 2860 .029.037 3-8 4.7 2775 .3 2860 .085-.O97 3-9 6.2 2790 .3 2870 .030 .033 3-10 6.5 2910 .3 2850 067-097 Cast without slag Table V11 shows the pipe wall thickness measurements at the ends and at 1 foot intervals along Pipe No. 3-9. The measurements are taken on each side of the cut in the pipe from the cold end to the hot end.
' TABLE vn WALL THICKNESS MEASUREMENTS OF 494-1NCH O.D. PIPE NO. 3-9. HEAT NO. 2681 Wall Thickness. Inch ness value difference of from one to two points from ID. to O.D., the OD. being slightly harder.
Chemical analysis was again taken from the [.D. and
0D. for a carbon check. Results of the 1D. and OD. carbon check showed little or no carbon difference, both areas being within the specification range.
It will be seen from the foregoing data that the use of molten slag within a centrifugally cast tube will markedly reduce the internal diameter shrinkage depth as well as improve the internal surface of the tube.
While we have illustrated and described certain presently preferred practices and embodiments of our invention in the foregoing specification, it will be understood that this invention may be otherwise embodied within the scope of the following claims.
1. The method of centrifugally casting hollow metal objects comprising the steps of:
a. casting molten metal intoa rotating centrifugal casting mold,
b. adding a molten self-disintegrating slag which on solid state transformation breaks up to said rotating centrifugal casting mold immediately after casting said molten metal therein,
c. cooling said mold to solidify the metal and cause the slag to solidify and disintegrate, and
d. removing the cast metal object from the mold and the disintegrated slag from the cast object.
is added within 5 seconds from the termination of pouring of molten metal.
6. The method of centrifugally casting hollow metal objects comprising the steps of:
a. casting molten metal into a rotating centrifugal casting mold,
b. adding a molten self-disintegrating slag containing dissolved carbon to said rotating centrifugal casting mold immediately after casting said molten metal therein,
c. cooling said mold to solidify the metal and cause the slag to solidify and disintegrate, and
d. removing the cast metal object from the mold and the disintegrated slag from the cast object, whereby the internal diameter of the metal is case hardened by the addition of carbon by including dissolved carbon in the slag at the time of addition.
7. The method as claimed in claim 6 wherein the carbon is dissolved in the slag by holding the slag in a carbon crucible at elevated temperature.