CA2105361C - Method of expendable pattern casting using sand with specific thermal properties - Google Patents
Method of expendable pattern casting using sand with specific thermal properties Download PDFInfo
- Publication number
- CA2105361C CA2105361C CA 2105361 CA2105361A CA2105361C CA 2105361 C CA2105361 C CA 2105361C CA 2105361 CA2105361 CA 2105361 CA 2105361 A CA2105361 A CA 2105361A CA 2105361 C CA2105361 C CA 2105361C
- Authority
- CA
- Canada
- Prior art keywords
- sand
- pattern
- silicon
- alloy
- aluminum
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22C—FOUNDRY MOULDING
- B22C1/00—Compositions of refractory mould or core materials; Grain structures thereof; Chemical or physical features in the formation or manufacture of moulds
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22C—FOUNDRY MOULDING
- B22C9/00—Moulds or cores; Moulding processes
- B22C9/02—Sand moulds or like moulds for shaped castings
- B22C9/04—Use of lost patterns
- B22C9/046—Use of patterns which are eliminated by the liquid metal in the mould
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Materials Engineering (AREA)
- Molds, Cores, And Manufacturing Methods Thereof (AREA)
Abstract
A method of producing a metal casting utilizing an expendable polymeric foam pattern along with unbonded sand having specific thermal properties. The pattern, formed of a material such as polystyrene, has a configuration corresponding to that of the article to be cast. The pattern is placed with an outer flask and unbonded sand surrounds the pattern as well as filling the cavities in the pattern. The sand has a linear expansion of less than 1% from 0°C to 1600°C, a particle size of 0.001 inch to 0.015 inch, and a heat diffusivity greater than 1500 J/m2/°K/s1/2. A molten metal, such as a hypoeutectic or hypereutectic aluminum silicon alloy or a ferrous alloy, is fed into the mold in contact with the pattern causing the pattern to vaporize with the vapor being entrapped within the interstices of the sand while the molten metal fills the space initially occupied by the foam pattern to produce a cast article. The physical properties of the sand enable articles to be cast having more precise tolerances.
Description
2~~~3~~ _ METHOD OF EXPENDABLE PATTERN CASTING
USING SAND WITH SPECIFIC THERMAL PROPERTIES
Backqround of the Invention Expendable Pattern casting, also known as lost foam casting, is a known casting technique in which a pattern formed of an polymeric foam material, such as polystyrene or polymethylmethacrylate, is supported in a flask and surrounded by an unbonded particulate material, such as silica sand. When the molten metal contacts the 20 pattern, the foam material decomposes with the products of decomposition passing into the interstices of the sand while the molten metal replaces the void formed by the expended foam material to produce a cast part which is identical in configuration to the pattern.
In the conventional expendable pattern casting process, the sand which surrounds the pattern and fills the cavities in the pattern is unbonded and free flowing and this differs from traditional sand casting processes, wherein the sand is utilized with various types of bind-ers. However, after compaction, the unbonded sand density is generally higher than the density of molds made with bonded sand, and therefore the rigidity or stiffness of compacted unbonded sand is not deficient relative to bonded sand molds.
Traditionally, silica sand has been used exclusively as the molding material in expendable pattern casting because it is readily available and inexpensive.
A conventional expendable pattern casting process is only capable of matching the precision of green sand casting and has riot been considered a precision sand casting process. This lack of precision for a process that uses metal molds to make the foam patterns, is a drawback of the process.
Aluminum silicon alloys have been cast utiliz-ing expendable pattern casting techniques as disclosed in U.S. Patent No. 4,966,220. Aluminum silicon alloys containing less than about 11.60 by weight of silicon are referred to as hypoeutectic alloys and the unmodified alloys have a microstructure consisting of primary alumi-num dendrites, with a eutectic composed of acicular silicon in an aluminum matrix. Hypoeutectic aluminum silicon alloys have seen extensive use in the past but lack wear resistance.
Hypereutectic aluminum silicon alloy, those containing more than about 11.6% silicon, contain primary silicon crystals which are precipitated as the alloy is cooled between the liquidus temperature and the eutectic temperature. nue to the high hardness of the precipitat-ed primary silicon crystals, these alloys have good wear resistance but are difficult to machine, a condition which limits their use as casting alloys.
Normally, a solid phase in a '°liquid plus solid" field has either a lower or higher density than the liquid phase, but almost never the same density. If the solid phase is less dense than the liquid phase, floatation of the solid phase will result. On the other hand, if the solid phase is more dense, a settling of the solid phase will occur. In either case, an increased or widened solidification range, which is a temperature range over-which an allay will solidify, will increase the time period for solidification and accentuate the phase separation. With a hypereutectic aluminum-silicon alloy, the silicon particles have a lesser density than the liquid phase so that the floatation condition pre-vails. Thus, as the solidification range is widened, the tendency fox floatation of Large primary silicon parti-cles increases, thus resulting in a less uniform distri-bution of -silicon particles in the cast alloy. Converse-ly, if the rate of cooling through the solidification range is increased, the tendency for floatation of the primary silicon particles is decreased resulting in a more uniform distribution of smaller silicon particles in the alloy. However, at sand casting cooling rates, ~~Q
USING SAND WITH SPECIFIC THERMAL PROPERTIES
Backqround of the Invention Expendable Pattern casting, also known as lost foam casting, is a known casting technique in which a pattern formed of an polymeric foam material, such as polystyrene or polymethylmethacrylate, is supported in a flask and surrounded by an unbonded particulate material, such as silica sand. When the molten metal contacts the 20 pattern, the foam material decomposes with the products of decomposition passing into the interstices of the sand while the molten metal replaces the void formed by the expended foam material to produce a cast part which is identical in configuration to the pattern.
In the conventional expendable pattern casting process, the sand which surrounds the pattern and fills the cavities in the pattern is unbonded and free flowing and this differs from traditional sand casting processes, wherein the sand is utilized with various types of bind-ers. However, after compaction, the unbonded sand density is generally higher than the density of molds made with bonded sand, and therefore the rigidity or stiffness of compacted unbonded sand is not deficient relative to bonded sand molds.
Traditionally, silica sand has been used exclusively as the molding material in expendable pattern casting because it is readily available and inexpensive.
A conventional expendable pattern casting process is only capable of matching the precision of green sand casting and has riot been considered a precision sand casting process. This lack of precision for a process that uses metal molds to make the foam patterns, is a drawback of the process.
Aluminum silicon alloys have been cast utiliz-ing expendable pattern casting techniques as disclosed in U.S. Patent No. 4,966,220. Aluminum silicon alloys containing less than about 11.60 by weight of silicon are referred to as hypoeutectic alloys and the unmodified alloys have a microstructure consisting of primary alumi-num dendrites, with a eutectic composed of acicular silicon in an aluminum matrix. Hypoeutectic aluminum silicon alloys have seen extensive use in the past but lack wear resistance.
Hypereutectic aluminum silicon alloy, those containing more than about 11.6% silicon, contain primary silicon crystals which are precipitated as the alloy is cooled between the liquidus temperature and the eutectic temperature. nue to the high hardness of the precipitat-ed primary silicon crystals, these alloys have good wear resistance but are difficult to machine, a condition which limits their use as casting alloys.
Normally, a solid phase in a '°liquid plus solid" field has either a lower or higher density than the liquid phase, but almost never the same density. If the solid phase is less dense than the liquid phase, floatation of the solid phase will result. On the other hand, if the solid phase is more dense, a settling of the solid phase will occur. In either case, an increased or widened solidification range, which is a temperature range over-which an allay will solidify, will increase the time period for solidification and accentuate the phase separation. With a hypereutectic aluminum-silicon alloy, the silicon particles have a lesser density than the liquid phase so that the floatation condition pre-vails. Thus, as the solidification range is widened, the tendency fox floatation of Large primary silicon parti-cles increases, thus resulting in a less uniform distri-bution of -silicon particles in the cast alloy. Converse-ly, if the rate of cooling through the solidification range is increased, the tendency for floatation of the primary silicon particles is decreased resulting in a more uniform distribution of smaller silicon particles in the alloy. However, at sand casting cooling rates, ~~Q
improvements in wear resistance or machinability by using different sand types, have not been recognized.
It is recognized in the casting art that using a molding material that extracts heat more rapidly from the molten metal and allows it to solidify at a faster rate, yields a casting with superior mechanical proper-ties. A cooling rate increase of three orders of magnitude (i.e. a 1000 times increase) decreases the dendritic arm spacing of the primary aluminum phase of hypoeutectic aluminum-silicon alloys by one order of magnitude (i.e. a factor of 10). This microstructure change results in an increase in mechanical properties.
Thus, castings produced using metal molds, which extract heat rapidly, generally exhibit superior mechanical properties as opposed to castings produced by sand casting or expendable pattern casting processes that utilize sand as a molding material. However, when using sand as a molding material, as in sand casting or expendable pattern casting, daubling the cooling rate (which is theoretically the most that can be expected from the higher heat diffusivity abtainable with any sand media), decreases the dendritic arm spacing of hypoeutectic aluminum-silicon alloys by approximately 10%
and this reduction results in only a 5% increase in the ultimate tensile strength. Thus sand casting properties of hypoeutectic aluminum-silicon alloys are never listed in the reference books by sand type.
One skilled in the metal casting art does not expect the temperature of the sand to have a significant influence on the dimensional size of castings produced by any of the sand casting processes. The major reason for this oversight is because, except for the expendable pattern casting process which uses unbonded sand, sand casting processes use bonded sand molds and these are used at the semi-uncontrolled ambient temperatures seen on 'the foundry floor. The economics of achieving through-put in the foundry and the cost of carrying an unnecessarily high inventory of molds on the foundry floor, dictate that the bonded sand molds be used in some orderly just-in-time approach. As a result, it is not the practice of foundries to heat or to cool the sand molds in a separate "conditioning" or stabilizing area and there has been no recognition that the temperature of the sand mold has a significant effect on the dimensional size or tolerances of the resulting castings that are produced with the molds. ' Summary of the Invention The invention is directed to a method of expendable pattern casting utilizing a sand molding material having specific physical properties to produce castings having more precise dimensions or tolerances.
The invention has particular application to the casting of engine blocks for internal combustion engines using not only hypoeutectic aluminum-silicon alloys containing about 4% to 11% by weight of silicon and hypereutectic aluminum- silicon alloys containing from about 16% to 30%
by weight of silicon, but other aluminum-silicon alloys between these silicon composition limits.
Tn the method of the invention, a polymeric foam pattern is produced having a configuration corre-sponding to the article to be cast. The foam pattern is supported in a flask and an unbonded sand is fed into the flask, surrounding the pattern and filling the cavities in the pattern.
The sand has a heat diffusivity greater than 1500 J/Mz/K/s~h, a linear expansion from 0°C to 1600°C of less than 1%, and a particle size in 'the range of 0.002 to 0.015 inch. Chromite sand, silicon carbide sand, olivine sand, and carbon sand have properties falling within these limits and are examples of sands which can be utilized.
When the foam pattern is contacted by 'the molten metal, the pattern will decompose and the products of decomposition will be entrapped within the interstices of the unbonded sand while the metal will fill the space initially occupied by the foam pattern, thereby producing cast article which corresponds in configuration to the a foam pattern.
The thermal properties of the sand allow heat to be extracted at a faster rate from the molten meta l and coupled with the much smaller expansion of the sand mold material, more precise castings are produced.
With the use of hypereutectic aluminum silicon l0 alloys, the thermal properties of the sand and the faster extraction of heat promotes enhanced under-cooling below the liquidus of the molten metal and increased nucleation of the primary silicon resulting in a smaller primary silicon particle size in the cast article, thereby improving the machinability of the casting.
As a further advantage, the use of the sand with the above specified properties produces a more uniform shrinkage of the cast metal on solidification, resulting in a coefficient of variation of shrinkage of less than 45%, as compared to a coefficient of variation of shrinkage of about 50% when using silica sand. The reduction in the coefficient produces a more precisely dimensional casting.
Other advantages will appear in the course of the following description.
Description of the Drawinas The drawings illustrate the best mode presently contemplated of carrying out the invention.
In the drawings:
Fig. 1 is a graph showing the linear expansion of various sands with temperature; and Fig. 2 is a graph showing the variation in dimensions of a three cylinder engine block when using silica sand at different temperatures.
It is recognized in the casting art that using a molding material that extracts heat more rapidly from the molten metal and allows it to solidify at a faster rate, yields a casting with superior mechanical proper-ties. A cooling rate increase of three orders of magnitude (i.e. a 1000 times increase) decreases the dendritic arm spacing of the primary aluminum phase of hypoeutectic aluminum-silicon alloys by one order of magnitude (i.e. a factor of 10). This microstructure change results in an increase in mechanical properties.
Thus, castings produced using metal molds, which extract heat rapidly, generally exhibit superior mechanical properties as opposed to castings produced by sand casting or expendable pattern casting processes that utilize sand as a molding material. However, when using sand as a molding material, as in sand casting or expendable pattern casting, daubling the cooling rate (which is theoretically the most that can be expected from the higher heat diffusivity abtainable with any sand media), decreases the dendritic arm spacing of hypoeutectic aluminum-silicon alloys by approximately 10%
and this reduction results in only a 5% increase in the ultimate tensile strength. Thus sand casting properties of hypoeutectic aluminum-silicon alloys are never listed in the reference books by sand type.
One skilled in the metal casting art does not expect the temperature of the sand to have a significant influence on the dimensional size of castings produced by any of the sand casting processes. The major reason for this oversight is because, except for the expendable pattern casting process which uses unbonded sand, sand casting processes use bonded sand molds and these are used at the semi-uncontrolled ambient temperatures seen on 'the foundry floor. The economics of achieving through-put in the foundry and the cost of carrying an unnecessarily high inventory of molds on the foundry floor, dictate that the bonded sand molds be used in some orderly just-in-time approach. As a result, it is not the practice of foundries to heat or to cool the sand molds in a separate "conditioning" or stabilizing area and there has been no recognition that the temperature of the sand mold has a significant effect on the dimensional size or tolerances of the resulting castings that are produced with the molds. ' Summary of the Invention The invention is directed to a method of expendable pattern casting utilizing a sand molding material having specific physical properties to produce castings having more precise dimensions or tolerances.
The invention has particular application to the casting of engine blocks for internal combustion engines using not only hypoeutectic aluminum-silicon alloys containing about 4% to 11% by weight of silicon and hypereutectic aluminum- silicon alloys containing from about 16% to 30%
by weight of silicon, but other aluminum-silicon alloys between these silicon composition limits.
Tn the method of the invention, a polymeric foam pattern is produced having a configuration corre-sponding to the article to be cast. The foam pattern is supported in a flask and an unbonded sand is fed into the flask, surrounding the pattern and filling the cavities in the pattern.
The sand has a heat diffusivity greater than 1500 J/Mz/K/s~h, a linear expansion from 0°C to 1600°C of less than 1%, and a particle size in 'the range of 0.002 to 0.015 inch. Chromite sand, silicon carbide sand, olivine sand, and carbon sand have properties falling within these limits and are examples of sands which can be utilized.
When the foam pattern is contacted by 'the molten metal, the pattern will decompose and the products of decomposition will be entrapped within the interstices of the unbonded sand while the metal will fill the space initially occupied by the foam pattern, thereby producing cast article which corresponds in configuration to the a foam pattern.
The thermal properties of the sand allow heat to be extracted at a faster rate from the molten meta l and coupled with the much smaller expansion of the sand mold material, more precise castings are produced.
With the use of hypereutectic aluminum silicon l0 alloys, the thermal properties of the sand and the faster extraction of heat promotes enhanced under-cooling below the liquidus of the molten metal and increased nucleation of the primary silicon resulting in a smaller primary silicon particle size in the cast article, thereby improving the machinability of the casting.
As a further advantage, the use of the sand with the above specified properties produces a more uniform shrinkage of the cast metal on solidification, resulting in a coefficient of variation of shrinkage of less than 45%, as compared to a coefficient of variation of shrinkage of about 50% when using silica sand. The reduction in the coefficient produces a more precisely dimensional casting.
Other advantages will appear in the course of the following description.
Description of the Drawinas The drawings illustrate the best mode presently contemplated of carrying out the invention.
In the drawings:
Fig. 1 is a graph showing the linear expansion of various sands with temperature; and Fig. 2 is a graph showing the variation in dimensions of a three cylinder engine block when using silica sand at different temperatures.
D~~eri~tienaof the Preferred Embodiment The invention relates to a method of expendable pattern casting utilizing unbonded sand having specific thermal properties as a molding material.
In carrying out the invention, a polymeric foam pattern is produced from a material such as polystyrene or polymethylmethacrylate to provide a pattern having a configuration corresponding to that of the article to be cast. The foam pattern itself is produced by 20 conventional procedures.
As in conventional expendable foam casting, the pattern can be coated with a porous ceramic material which acts to prevent a metal/sand reaction and facili-tates cleaning of the cast metal part. The ceramic coating is normally applied by immersing the pattern in a bath of ceramic wash, draining the excess wash from the pattern and drying the wash to provide the porous ceramic coating.
The process of the invention can be used with any desired metal .or alloy and has particular application in casting ferrous metals, such as cast iron or steel, or aluminum-silicon alloys, either hypoeutectic or hyper-eutectic aluminum-silicon alloys. In general, the hypereutectic aluminum silicon alloys contain by weight 12o to 30% silicon, 0.4o to S.Oo magnesium, up to 0.30 manganese, up to 1.4o iron, up to 5.0% copper, and the balance aluminum.
Specific examples of hypereutectic aluminum silicon alloys to be used are as follows in weight per-cent:
Silicon 16.90%
Iron 0.920 Copper 0.14%
Manganese 0.12%
Magnesium 0.41%
Aluminum 81.51%
~1~~3~~.
_ 7 _ Silicon 20.10%
Iron 0.20%
Copper 0..33%
Manganese 0.18%
Magnesium o..71%
Aluminum 78.400 The hypoeutectic aluminum-silicon alloys contain by weight less than 12% silicon, and one common sand casting alloy contains from 6.5% to 7.5% by weight of silicon, 0.25% to 0.450 by weight of magnesium, up to 0.6% iron, up to 0.2% copper, up to 0.25% titanium, up to 0.35% zinc, up to 0.350 manganese, and the balance aluminum. Another common hypoeutectic aluminum--silicon alloy that can be used in the invention contains from 5.5% to 6.5% by weight of silicon, from 3.0% to 4.0% by weight of copper, from 0.1% to 0.5% by weight of magnesium, up to 1.2% iron, up to 0.8% manganese, up to 0.5% nickel, up to 3.0% zinc, up to 0.25% titanium, and the balance aluminum.
Specific examples of hypoeutectic~aluminum silicon alloys to be used are as follows in weight percent:
Silicon 7.10%
Magnesium 0.31%
Copper 0.05%
Titanium 0.05%
Zinc 0.100 Manganese 0.05%
Aluminum 92.210 210~3~~.
Silicon 6.210 Copper 3.150 Magnesium 0..32%
Iron 0.800 Manganese 0.520 Nickel 0.34%
Zinc 1.02%
Titanium 0.20%
Aluminum 87.35%
Traditionally, silica sand has been used as the molding material in expendable pattern casting due to the fact that silica sand is readily available and is inexpensive. Through the development of the invention, it has been discovered that the use of silica sand pres-ents certain drawbacks when utilized in expendable patt-ern casting procedures that were heretofore unrecognized, and it has been further discovered that the unbonded sand molding material should have certain physical properties, not obtainable with silicon sand, in order to achieve precision castings.
First, it has been found that the sand temperature at the start of pouring in the expendable pattern casting process has a significant effect on the dimensional size of the casting produced in the process.
Secondly, it has been found that the co-efficient of variation of shrinkage of the metal casting is significantly improved if the sand molding material has a linear expansion of less than 10 Thirdly, it has further been found that the shrinkage value of the unbonded sand should nearly match the' unconstrained shrinkage value of the cast metal, unlike bonded sand castings with large cores, which exhibit unpredictable lower shrinkage values, as compared to the shrinkage value of the cast metal.
And fourthly, it has also been discovered that when dealing with hypereutectic aluminum-silicon _ g alloys, primary silicon particle size is primarily affected by the initial cooling rate just below the liquidus (for which sand type has an influence) rather than by a faster average cooling rate through the entire liquid plus solid solidification range and heavily influenced by a high eutectic volume fraction.
The physical properties of sand, particularly the thermal properties, greatly effect the precision of casting when using expendable foam patterns. To provide the improved precision in casting, the sand should have a heat diffusivity greater than 1500 J/MZ/°K/s'~, a total linear expansion from 0°C to 1600°C of less than 1%, and a particle size in the range of 0.001 inch to 0.015 inch.
Chromite sand (FeCr204), silicon carbide sand, carbon sand, and olivine sand (a solid solution of forsterite, Mg2Si04, and fayalite, Fe2Si04) are examples of sands that can be used in the process of the invention.
A comparison of the physical properties of chromite sand, silicon carbide sand and silica sand are shown in the following table.
TABLE I
Silicon Silica Chromite Carbide Sand Sand Sand Thermal conductivity 0.90-0.61 1.09 3.25 (watts/m/°K) Density (Kg/m3) 1500 2400 2000 Specific heat (J/Kg/°K) 1130-1172 963 840 Thermal diffusivity 0.360-0.512 0.472 2.0 (MZ/s x 10'6) Heat diffusivity 1017-1258 1587 2340 (J/MZ/°KIs~~) The thermal conductivity of a material is the quantity of heat which flows per unit time through a unit area of a mass of the material of unit thickness when there is a difference of 1° in the temperatures across opposite faces of the mass. The time rate of change of r ~~.~~36~.
the temperature, at any location is proportional to the instantaneous slope of temperature gradient. The propor-tionality constant is called the thermal diffusivity and is defined as the thermal conductivity divided by the volumetric heat capacity where the volumetric heat capac-ity is the heat per unit volume necessary to raise the temperature of the mass 1°.
The heat diffusivity, on 'the other hand, is a measure of the rate at which the mold can absorb heat and is the square root of the product of the thermal conduc tivity, the density and the specific heat. As such, heat diffusivity is directly related to solidification rate of the molten metal.
It has been found that the linear expansion of the sand with temperature is an important factor in providing precise castings, and the linear expansion of the sand should be less than 10 over a temperature range of 0°C to 1600°C, and preferably less than 0.750, over a temperature range of 0°C to 700°C. Figure 1 is a graph showing the change in linear expansion of silica sand, chromite sand and olivine sand with temperature. The curve of silica sand shows a substantial increase in expansion as the temperature of the silica sand approach-es approximately 550°C. From the above graph, it is noted that chromite and olivine do not undergo a similar abrupt expansion as does the silica sand.
The importance of the linear expansion is apparent when one compares the thermal diffusivity of a casting metal, such as an aluminum alloy, with that of sand. The thermal diffusivity of an aluminum alloy is approximately 6.2 x 10'5 m'/s which is approximately 150 times greater than the thermal diffusivity of the sands as shown in Table T above. This means that the average distance through which heat flows in a given time is approximately 12 times greater for the aluminum alloy than for sand, resulting in a heat build up at the sand/
metal interface which causes the sand mold cavity to expand. Since the thermal expansion coefficient of silica sand is approximately 4 times greater than that of chromite sand, any temperature increase at the metal/sand interface will cause the silica sand to expand substanti-ally more than chromite sand and therefore will produce a larger dimensional casting. Also, since the molten metal/sand interface has moved outward before the start of solidification, the calculated shrinkage value obtain-ed on the larger casting will result in an apparent lower (and unpredictable) shrinkage value for the solidified metal.
As noted above, the heat diffusivity of the sand is directly related to the solidification rate of the molten metal. From the heat diffusivity data shown in Table I above, it is seen that the use of chromite sand should increase the solidification rate of the metal, i.e. the time required to pass between the liquidus and solidus temperatures, over that using silica sand by approximately 26% to 56o due to the greater heat diffusivity of the chromite sand. This improvement in the solidification rate in itself may not be seen as a worthwhile economic advantage but when considered with the large expansion that occurs with silica sand at about 550°C, a substantial improvement in the precision of the castings is achieved.
Furthermore, an unexpected enhanced nucleat-ing phenomena is obtained when utilizing hypereutectic aluminum silicon alloys in the process of the invention.
With hypereutectic aluminum silicon alloys, primary silicon crystals are precipitated as the alloy is cooled from solution temperature. When hypereutectic aluminum silicon alloys are cast in the process of the invention using the sand of the above-noted thermal properties, heat is more readily extracted from the molten metal (before heat saturation occurs at the molten metal/sand interface) which contributes to enhanced undercooling below the liquidus temperature of the alloy which in turn promotes increased nucleation of the primary silicon resulting in a smaller silicon particle size in the cast article. The reduction in particle size of the silicon improves the machinability of the alloy making the cast alloy more valuable for articles such as engine blocks.
From the data in Table I, and a comparison of the heat diffusivity values, the use of chromire sand produces about a 27o increase in the solidification rate, 20 as opposed to the use of silica sand. With hypoeutectic aluminum-silicon alloys a 27% increase in the solidifica-tion rate results in an insignificant improvement in mechanical properties (i.e. decrease in the primary aluminum dendrite arm spacing). However, with hyper-eutectic aluminum-silicon alloys the primary silicon particle size appears to be very sensitive to a marginal increase of 27o in the solidificai:ion rate. The reason for this different sensitivity is because different fundamental mechanisms are operating. The mechanical properties of hypoeutectic aluminum-silicon alloys are controlled by the dendrite arm spacing of the primary aluminum and this spacing is controlled by a growth mechanism dictated by the solidification rate. The silicon particle size of hypereutectic aluminum-silicon alloys is controlled by a nucleation mechanism and 'this in turn is controlled by the character of the under-cooling immediately below the liquidus.
A consideration of the microstructure difference between hypoeutectic and hypereutectic aluminum-silicon allays is quite helpful in illustrating how the heat of fusion is dissipated during solidifica-tion in these alloys. At a temperature slightly above the eutectic temperature, a hypoeutectic aluminum alloy of Example 3 (microstructure: 40% primary aluminum, 60%
eutectic liquid) and a hypereutectic aluminum-silicon alloy of Example 1 (microstructure: 10o primary silicon, 90o eutectic liquid) both have given up approximately an ~1~~351 equal amount of heat to the sand molding material, since the heat of fusion of silicon is 4.5 times that of aluminum. However, as the remaining eutectic liquid solidifies, the hypereutectic alloy gives up increment-s ally 50% more heat on a volume basis (since it contains 50% more eutectic liquid) than the hypoeutectic alloy (38% on an overall basis), which inherently slows the solidification of the hypereutectic alloy compared to the hypoeutectic alloy. The insight into obtaining a smaller primary silicon particle size in the hypereutectic alloy is to focus on the temperature range immediately below the liquidus temperature where the nucleation phenomena can be affected by an incremental faster cooling rate and not t4 focus on the temperature where silicon rejection is being accommodated by the growth of existing primary silicon particles. Thus, the use of chromite sand in the expendable pattern casting process, which can affect a 26-56% increase in the solidification rate (which shows a parabolic dependence with time and is therefore more effective in "early" time rather than in "late" time) has a most significant effect on primary silicon particle size, and this has heretofore not been recognized.
By comparison, die casting with large metal molds that function as very large heat sinks, overwhelms the nucleation at temperatures immediately blow the liquidus temperature, as well as at temperatures immedi-ately below the eutectic.temperature. As a result, the die cast microstructure for a hypereutectic aluminum-silicon alloy, even containing no phosphorous additions, consists of a refined primary silicon as well as a refined eutectic silicon.
The microstructure of an expendable pattern cast hypereutectic aluminum-silicon alloy, by contrast, does not contain refined eutectic silicon in the microstructure because the cooling rate is far too slow.
In fact, primary silicon refinement in a hypereutectic aluminum-silicon alloy requires phosphorous additions in 2~.053~~.
all casting processes except a die casting process. In Gruzleski and Closset, "The Treatment of Liquid Aluminum-Silicon Alloys" (American Foundrymen's Society, Inc., 1990) it is stated that "hypereutectic alloys such as 390 are very difficult, if not impossible, to sand cast. Even with phosphorous treatment, solidification rates are so slow that unacceptably large primary phase particles form and float to the upper surfaces of the casting." Therefore, it has been believed that the cooling rate has an effect on the primary silicon particle size in a phosphorous treated alloy, but that the cooling rate effect refers to the entire liquid-solid range, as in die casting, and not just to the upper portion of the liquid-solid range. It is believed that the reason for this erroneous insight is that hyper-eutectic aluminum-silicon alloys have not been considered viable sand casting alloys and, therefore, sand casting developments with hypereutectic aluminum-silicon alloys have nct been investigated for commercial use, and further primary silicon particle size has never been studied, as a function of sand type, in any of the various sand casting processes The reduction in particle size of the silicon crystals can be illustrated by a comparison of casting a hypereutectic aluminum silicon alloy in an expendable pattern process using silica sand as a molding material as compared to using chromite sand. In this comparison, the aluminum silicon alloy contained lB.Oo silicon, 0.69%
magnesium, 0.1% copper, and the balance aluminum. The molten aluminum silicon alloy was poured at a temperature of 704°C (1300°F) into a flask containing silica sand at 26.7°C (80°F) and into a second chromite sand flask at 26.7°C (80°F), both containing a polystyrene sprue with three polystyrene foam patterns of a 60 horsepower 3 cylinder engine block connected to the sprue. Differenc-es in the primary silicon particle size were measured in the cast engine blocks with the two sand types. The average primary silicon particle size obtained by measur-ing 849 silicon particles utilizing silica sand in the casting was 30 microns with a coefficient of variation of primary silicon particle size of 500. The average prim-ary silicon particle size obtained by measuring 442 silicon particles produced using the chromite sand was 21.4 microns and a coefficient of variation in that average of 37%. Thus, the use of silica sand gave an average primary silicon particle size which was 39%
larger than that obtained through use of chromite sand and the coefficient of variation of the particle size using silica sand was substantially greater than that obtained with the chromite sand. In general, the average primary silicon particle size of the cast hypereutectic aluminum-silicon alloy produced by the invention is less than 30 microns and the coefficient of variation of particle size is less than 500 This test evidences the unexpected reduction in silicon particle size in a hypereutectic aluminum silicon alloy that. is achieved when using the specified sand in an expendable pattern casting process.
When casting an engine block for an internal combustion engine, the pattern is formed with a plurality of cylindrical bores which correspond to the cylinders in the cast block. In the flask the sand nat only surrounds the pattern, but also fills the bores thus providing sand cores. During casting, the molten metal will shrink as it solidifies. If the sand core does not "give" as the metal solidifies and shrinks around it, stresses can be set up in the casting and unpredictable diameters will be obtained in the cylinder bores. Thus, the sand used as the core should permit the core to follow the shrinkage of the solidifying metal.
The following Table summarizes repeat measurements of the average of twenty-five different critical dimensions of a complex 60 ~iP, three-cylinder, marine engine block using various sand molding materials in an expendable foam casting process and using the aluminum-silicon alloy of Example 3, above. The results show that when using chromite sand, silicon carbide sand, or carbon sand the shrinkage of the alloy has nearly matched the unconstricted contraction of the alloy as reported. in the literature. This is quite surprising, because complex engine blocks with large cores and produced in a sand casting process using bonded sand, generally exhibit smaller contraction results than the unconstricted contraction of the alloy. The different contraction results in the alloy, as shown in Table II, are believed to be the result of different degrees of constraint by the sand mold (and core) during cooling.
It is recognized that the hardness of ramming of the sand and the percentage of binder in chemically bonded sands significantly affects the contraction. Based on the above, the unbonded sand in the expendable foam casting process can be viewed as inherently offering less of a constraint during cooling than bonded sand and, therefore more sensitive to the phenomena of the expansion of the mold fram molten metals of higher heat content and/or from starting the casting process with heated sand. This latter factor is clearly reflected by the shrinkage values in the alloy of 0.00925 inch per inch and 0.007 inch per inch for 80°F and 160°F silicon sand, respectively. The larger dimensioned cast engine blocks resulting from the.use of heated silica sand simply reflect the larger expansion of the sand mold as a result of the higher sand temperature.
TABLE II
Average Coefficient Sand Sand Shrinkage of Variation Type Temperature (in Lin) of Shrinkage S7.liCd 160F 0.0070 600 SllICa 80F 0.0093 500 Chromite 80F 0.0119 35%
Silicon 80F 0.0110 37%
carbide Spherical 80F 0.0106 360 Carbon sand From the above table, it can be seen.that the use of chromite sand, silicon carbide sand, and carbon sand, resulted in a greater metal shrinkage rate in inches per inch than when using silica sand thus enabling an unbonded sand Core to more Closely follow the shrink-age of the alloy.
Equally important is the fact that the use of Chromite sand, silicon carbide sand and Carbon sand produced a substantially lower coefficient of variation of shrinkage in the metal casting, as compared to the use of silica sand. This means that the shrinkage at the various locations of measurement was more uniform and had less variance than that measured when using silica sand.
Repeat measurements on a cast 250 HP, V-6, 3 liter, marine engine block gave similar results. The ambient temperature, 80°F, silica sand yielded a shrink-age value of 0.0094 inch per inch, and the ambient temperature chromite sand yielded a shrinkage of 0.0118 inch per inch. In addition, the ambient temperature silica sand gave a precision, reflected by the coefficient of variation (of shrinkage), more than 500 less than that obtained through use of the Chromite sand.
These results further evidence that silica sand, as the molding material, produces larger dimensional engine blocks than the use of chromite sand. Moreover, the precision obtained with silica sand is significantly less ~~.~~3~1 than the precision obtained with chromite sand. The test results also indicate that the geometry differences between a V-6 engine block and an in-line three--cylinder block do not materially affect the shrinkage values obtained for the two different sand types.
The particle size of the sand should be in the range of 0.001 inch to 0.015 inch. An expendable pattern casting process requires a relatively coarse sand because of the sand permeability requirements and metal fill time requirements which are interrelated. In expendable pattern casting, the molten metal cannot completely fill the sand mold cavity until the products of foam decomposition have entered the interstices between the sand grains.
Not only are the thermal properties of the sand important in providing precision castings, but it has also been found that the temperature of the sand influences the casting. For example, in winter candi-tions in the foundry, the sand temperature may be in the range of 18.3°C (65°F) to 29.4°C (85°F). In summer, where the ambient temperature may be up to 32.2°C (90°F) or higher, the sand temperature can be in the range of 29.4°C (85°F) to 40.5°C (105°F). With the higher sand temperature in summer, the castings will have a somewhat larger dimension than castings produced in the winter with the sand at a lower temperature. Therefore, to compensate for this differential in dimensions in the cast part, the size of the expendable foam patterns can be adjusted. The dimension of the pattern can be changed by aging the plastic beads before molding, or by aging the molded parts after molding, or by selecting another foam bead type. Thus, by proper aging or selection of the beads, a larger pattern can be obtained which can be used in the winter to compensate for the lower sand temperature, thus resulting in cast parts which have the same dimensions regardless of the temperature of the sand.
Fig. 2 further illustrates the importance of the sand temperature on the precision of casting. Fig. 2 is a curve showing average measurements of an engine block dimension in inches as a function of the tempera-s Lure of unbonded silica sand used in an expendable pattern casting process. The engine block was cast from a hypoeutectic aluminum-silicon alloy having the composition of Example 3 above. As seen in Fig. 2, the average engine block dimension when using sand at ambient temperature of 80°F was 9.53 inches. As the sand temperature was increased to 160°F, the average block dimension also increased. to a value of about 9:59 inches, or an increase of 0.06 inch.
While the above curve shows the difference in dimensions obtained by using silica sand at various tempertures, similar expansion results, although smaller by a factor of approximately four, are obtained using chromite sand, silicon carbide sand, or carbon sand, thus indicating that sand temperature is a factor in obtaining precisely dimensioned castings.
' The invention is based on the discovery that more precise castings can be produced in an expendable pattern casting process by utilizing sand having specific thermal properties and cntrolling the sand temperature or correlating the sand temperature with the pattern size.
When dealing with hypereutectic aluminum silicon alloys, the invention provides a second and unexpected advantage in that the particle size of the precipitated silicon crystals is reduced which substantially improves the 30~ machinability of the alloy.
Various modes of carrying out the invention are contemplated as being within the scope of the following claims particularly pointing out and distinctly claiming the subject matter which is regarded as the invention.
In carrying out the invention, a polymeric foam pattern is produced from a material such as polystyrene or polymethylmethacrylate to provide a pattern having a configuration corresponding to that of the article to be cast. The foam pattern itself is produced by 20 conventional procedures.
As in conventional expendable foam casting, the pattern can be coated with a porous ceramic material which acts to prevent a metal/sand reaction and facili-tates cleaning of the cast metal part. The ceramic coating is normally applied by immersing the pattern in a bath of ceramic wash, draining the excess wash from the pattern and drying the wash to provide the porous ceramic coating.
The process of the invention can be used with any desired metal .or alloy and has particular application in casting ferrous metals, such as cast iron or steel, or aluminum-silicon alloys, either hypoeutectic or hyper-eutectic aluminum-silicon alloys. In general, the hypereutectic aluminum silicon alloys contain by weight 12o to 30% silicon, 0.4o to S.Oo magnesium, up to 0.30 manganese, up to 1.4o iron, up to 5.0% copper, and the balance aluminum.
Specific examples of hypereutectic aluminum silicon alloys to be used are as follows in weight per-cent:
Silicon 16.90%
Iron 0.920 Copper 0.14%
Manganese 0.12%
Magnesium 0.41%
Aluminum 81.51%
~1~~3~~.
_ 7 _ Silicon 20.10%
Iron 0.20%
Copper 0..33%
Manganese 0.18%
Magnesium o..71%
Aluminum 78.400 The hypoeutectic aluminum-silicon alloys contain by weight less than 12% silicon, and one common sand casting alloy contains from 6.5% to 7.5% by weight of silicon, 0.25% to 0.450 by weight of magnesium, up to 0.6% iron, up to 0.2% copper, up to 0.25% titanium, up to 0.35% zinc, up to 0.350 manganese, and the balance aluminum. Another common hypoeutectic aluminum--silicon alloy that can be used in the invention contains from 5.5% to 6.5% by weight of silicon, from 3.0% to 4.0% by weight of copper, from 0.1% to 0.5% by weight of magnesium, up to 1.2% iron, up to 0.8% manganese, up to 0.5% nickel, up to 3.0% zinc, up to 0.25% titanium, and the balance aluminum.
Specific examples of hypoeutectic~aluminum silicon alloys to be used are as follows in weight percent:
Silicon 7.10%
Magnesium 0.31%
Copper 0.05%
Titanium 0.05%
Zinc 0.100 Manganese 0.05%
Aluminum 92.210 210~3~~.
Silicon 6.210 Copper 3.150 Magnesium 0..32%
Iron 0.800 Manganese 0.520 Nickel 0.34%
Zinc 1.02%
Titanium 0.20%
Aluminum 87.35%
Traditionally, silica sand has been used as the molding material in expendable pattern casting due to the fact that silica sand is readily available and is inexpensive. Through the development of the invention, it has been discovered that the use of silica sand pres-ents certain drawbacks when utilized in expendable patt-ern casting procedures that were heretofore unrecognized, and it has been further discovered that the unbonded sand molding material should have certain physical properties, not obtainable with silicon sand, in order to achieve precision castings.
First, it has been found that the sand temperature at the start of pouring in the expendable pattern casting process has a significant effect on the dimensional size of the casting produced in the process.
Secondly, it has been found that the co-efficient of variation of shrinkage of the metal casting is significantly improved if the sand molding material has a linear expansion of less than 10 Thirdly, it has further been found that the shrinkage value of the unbonded sand should nearly match the' unconstrained shrinkage value of the cast metal, unlike bonded sand castings with large cores, which exhibit unpredictable lower shrinkage values, as compared to the shrinkage value of the cast metal.
And fourthly, it has also been discovered that when dealing with hypereutectic aluminum-silicon _ g alloys, primary silicon particle size is primarily affected by the initial cooling rate just below the liquidus (for which sand type has an influence) rather than by a faster average cooling rate through the entire liquid plus solid solidification range and heavily influenced by a high eutectic volume fraction.
The physical properties of sand, particularly the thermal properties, greatly effect the precision of casting when using expendable foam patterns. To provide the improved precision in casting, the sand should have a heat diffusivity greater than 1500 J/MZ/°K/s'~, a total linear expansion from 0°C to 1600°C of less than 1%, and a particle size in the range of 0.001 inch to 0.015 inch.
Chromite sand (FeCr204), silicon carbide sand, carbon sand, and olivine sand (a solid solution of forsterite, Mg2Si04, and fayalite, Fe2Si04) are examples of sands that can be used in the process of the invention.
A comparison of the physical properties of chromite sand, silicon carbide sand and silica sand are shown in the following table.
TABLE I
Silicon Silica Chromite Carbide Sand Sand Sand Thermal conductivity 0.90-0.61 1.09 3.25 (watts/m/°K) Density (Kg/m3) 1500 2400 2000 Specific heat (J/Kg/°K) 1130-1172 963 840 Thermal diffusivity 0.360-0.512 0.472 2.0 (MZ/s x 10'6) Heat diffusivity 1017-1258 1587 2340 (J/MZ/°KIs~~) The thermal conductivity of a material is the quantity of heat which flows per unit time through a unit area of a mass of the material of unit thickness when there is a difference of 1° in the temperatures across opposite faces of the mass. The time rate of change of r ~~.~~36~.
the temperature, at any location is proportional to the instantaneous slope of temperature gradient. The propor-tionality constant is called the thermal diffusivity and is defined as the thermal conductivity divided by the volumetric heat capacity where the volumetric heat capac-ity is the heat per unit volume necessary to raise the temperature of the mass 1°.
The heat diffusivity, on 'the other hand, is a measure of the rate at which the mold can absorb heat and is the square root of the product of the thermal conduc tivity, the density and the specific heat. As such, heat diffusivity is directly related to solidification rate of the molten metal.
It has been found that the linear expansion of the sand with temperature is an important factor in providing precise castings, and the linear expansion of the sand should be less than 10 over a temperature range of 0°C to 1600°C, and preferably less than 0.750, over a temperature range of 0°C to 700°C. Figure 1 is a graph showing the change in linear expansion of silica sand, chromite sand and olivine sand with temperature. The curve of silica sand shows a substantial increase in expansion as the temperature of the silica sand approach-es approximately 550°C. From the above graph, it is noted that chromite and olivine do not undergo a similar abrupt expansion as does the silica sand.
The importance of the linear expansion is apparent when one compares the thermal diffusivity of a casting metal, such as an aluminum alloy, with that of sand. The thermal diffusivity of an aluminum alloy is approximately 6.2 x 10'5 m'/s which is approximately 150 times greater than the thermal diffusivity of the sands as shown in Table T above. This means that the average distance through which heat flows in a given time is approximately 12 times greater for the aluminum alloy than for sand, resulting in a heat build up at the sand/
metal interface which causes the sand mold cavity to expand. Since the thermal expansion coefficient of silica sand is approximately 4 times greater than that of chromite sand, any temperature increase at the metal/sand interface will cause the silica sand to expand substanti-ally more than chromite sand and therefore will produce a larger dimensional casting. Also, since the molten metal/sand interface has moved outward before the start of solidification, the calculated shrinkage value obtain-ed on the larger casting will result in an apparent lower (and unpredictable) shrinkage value for the solidified metal.
As noted above, the heat diffusivity of the sand is directly related to the solidification rate of the molten metal. From the heat diffusivity data shown in Table I above, it is seen that the use of chromite sand should increase the solidification rate of the metal, i.e. the time required to pass between the liquidus and solidus temperatures, over that using silica sand by approximately 26% to 56o due to the greater heat diffusivity of the chromite sand. This improvement in the solidification rate in itself may not be seen as a worthwhile economic advantage but when considered with the large expansion that occurs with silica sand at about 550°C, a substantial improvement in the precision of the castings is achieved.
Furthermore, an unexpected enhanced nucleat-ing phenomena is obtained when utilizing hypereutectic aluminum silicon alloys in the process of the invention.
With hypereutectic aluminum silicon alloys, primary silicon crystals are precipitated as the alloy is cooled from solution temperature. When hypereutectic aluminum silicon alloys are cast in the process of the invention using the sand of the above-noted thermal properties, heat is more readily extracted from the molten metal (before heat saturation occurs at the molten metal/sand interface) which contributes to enhanced undercooling below the liquidus temperature of the alloy which in turn promotes increased nucleation of the primary silicon resulting in a smaller silicon particle size in the cast article. The reduction in particle size of the silicon improves the machinability of the alloy making the cast alloy more valuable for articles such as engine blocks.
From the data in Table I, and a comparison of the heat diffusivity values, the use of chromire sand produces about a 27o increase in the solidification rate, 20 as opposed to the use of silica sand. With hypoeutectic aluminum-silicon alloys a 27% increase in the solidifica-tion rate results in an insignificant improvement in mechanical properties (i.e. decrease in the primary aluminum dendrite arm spacing). However, with hyper-eutectic aluminum-silicon alloys the primary silicon particle size appears to be very sensitive to a marginal increase of 27o in the solidificai:ion rate. The reason for this different sensitivity is because different fundamental mechanisms are operating. The mechanical properties of hypoeutectic aluminum-silicon alloys are controlled by the dendrite arm spacing of the primary aluminum and this spacing is controlled by a growth mechanism dictated by the solidification rate. The silicon particle size of hypereutectic aluminum-silicon alloys is controlled by a nucleation mechanism and 'this in turn is controlled by the character of the under-cooling immediately below the liquidus.
A consideration of the microstructure difference between hypoeutectic and hypereutectic aluminum-silicon allays is quite helpful in illustrating how the heat of fusion is dissipated during solidifica-tion in these alloys. At a temperature slightly above the eutectic temperature, a hypoeutectic aluminum alloy of Example 3 (microstructure: 40% primary aluminum, 60%
eutectic liquid) and a hypereutectic aluminum-silicon alloy of Example 1 (microstructure: 10o primary silicon, 90o eutectic liquid) both have given up approximately an ~1~~351 equal amount of heat to the sand molding material, since the heat of fusion of silicon is 4.5 times that of aluminum. However, as the remaining eutectic liquid solidifies, the hypereutectic alloy gives up increment-s ally 50% more heat on a volume basis (since it contains 50% more eutectic liquid) than the hypoeutectic alloy (38% on an overall basis), which inherently slows the solidification of the hypereutectic alloy compared to the hypoeutectic alloy. The insight into obtaining a smaller primary silicon particle size in the hypereutectic alloy is to focus on the temperature range immediately below the liquidus temperature where the nucleation phenomena can be affected by an incremental faster cooling rate and not t4 focus on the temperature where silicon rejection is being accommodated by the growth of existing primary silicon particles. Thus, the use of chromite sand in the expendable pattern casting process, which can affect a 26-56% increase in the solidification rate (which shows a parabolic dependence with time and is therefore more effective in "early" time rather than in "late" time) has a most significant effect on primary silicon particle size, and this has heretofore not been recognized.
By comparison, die casting with large metal molds that function as very large heat sinks, overwhelms the nucleation at temperatures immediately blow the liquidus temperature, as well as at temperatures immedi-ately below the eutectic.temperature. As a result, the die cast microstructure for a hypereutectic aluminum-silicon alloy, even containing no phosphorous additions, consists of a refined primary silicon as well as a refined eutectic silicon.
The microstructure of an expendable pattern cast hypereutectic aluminum-silicon alloy, by contrast, does not contain refined eutectic silicon in the microstructure because the cooling rate is far too slow.
In fact, primary silicon refinement in a hypereutectic aluminum-silicon alloy requires phosphorous additions in 2~.053~~.
all casting processes except a die casting process. In Gruzleski and Closset, "The Treatment of Liquid Aluminum-Silicon Alloys" (American Foundrymen's Society, Inc., 1990) it is stated that "hypereutectic alloys such as 390 are very difficult, if not impossible, to sand cast. Even with phosphorous treatment, solidification rates are so slow that unacceptably large primary phase particles form and float to the upper surfaces of the casting." Therefore, it has been believed that the cooling rate has an effect on the primary silicon particle size in a phosphorous treated alloy, but that the cooling rate effect refers to the entire liquid-solid range, as in die casting, and not just to the upper portion of the liquid-solid range. It is believed that the reason for this erroneous insight is that hyper-eutectic aluminum-silicon alloys have not been considered viable sand casting alloys and, therefore, sand casting developments with hypereutectic aluminum-silicon alloys have nct been investigated for commercial use, and further primary silicon particle size has never been studied, as a function of sand type, in any of the various sand casting processes The reduction in particle size of the silicon crystals can be illustrated by a comparison of casting a hypereutectic aluminum silicon alloy in an expendable pattern process using silica sand as a molding material as compared to using chromite sand. In this comparison, the aluminum silicon alloy contained lB.Oo silicon, 0.69%
magnesium, 0.1% copper, and the balance aluminum. The molten aluminum silicon alloy was poured at a temperature of 704°C (1300°F) into a flask containing silica sand at 26.7°C (80°F) and into a second chromite sand flask at 26.7°C (80°F), both containing a polystyrene sprue with three polystyrene foam patterns of a 60 horsepower 3 cylinder engine block connected to the sprue. Differenc-es in the primary silicon particle size were measured in the cast engine blocks with the two sand types. The average primary silicon particle size obtained by measur-ing 849 silicon particles utilizing silica sand in the casting was 30 microns with a coefficient of variation of primary silicon particle size of 500. The average prim-ary silicon particle size obtained by measuring 442 silicon particles produced using the chromite sand was 21.4 microns and a coefficient of variation in that average of 37%. Thus, the use of silica sand gave an average primary silicon particle size which was 39%
larger than that obtained through use of chromite sand and the coefficient of variation of the particle size using silica sand was substantially greater than that obtained with the chromite sand. In general, the average primary silicon particle size of the cast hypereutectic aluminum-silicon alloy produced by the invention is less than 30 microns and the coefficient of variation of particle size is less than 500 This test evidences the unexpected reduction in silicon particle size in a hypereutectic aluminum silicon alloy that. is achieved when using the specified sand in an expendable pattern casting process.
When casting an engine block for an internal combustion engine, the pattern is formed with a plurality of cylindrical bores which correspond to the cylinders in the cast block. In the flask the sand nat only surrounds the pattern, but also fills the bores thus providing sand cores. During casting, the molten metal will shrink as it solidifies. If the sand core does not "give" as the metal solidifies and shrinks around it, stresses can be set up in the casting and unpredictable diameters will be obtained in the cylinder bores. Thus, the sand used as the core should permit the core to follow the shrinkage of the solidifying metal.
The following Table summarizes repeat measurements of the average of twenty-five different critical dimensions of a complex 60 ~iP, three-cylinder, marine engine block using various sand molding materials in an expendable foam casting process and using the aluminum-silicon alloy of Example 3, above. The results show that when using chromite sand, silicon carbide sand, or carbon sand the shrinkage of the alloy has nearly matched the unconstricted contraction of the alloy as reported. in the literature. This is quite surprising, because complex engine blocks with large cores and produced in a sand casting process using bonded sand, generally exhibit smaller contraction results than the unconstricted contraction of the alloy. The different contraction results in the alloy, as shown in Table II, are believed to be the result of different degrees of constraint by the sand mold (and core) during cooling.
It is recognized that the hardness of ramming of the sand and the percentage of binder in chemically bonded sands significantly affects the contraction. Based on the above, the unbonded sand in the expendable foam casting process can be viewed as inherently offering less of a constraint during cooling than bonded sand and, therefore more sensitive to the phenomena of the expansion of the mold fram molten metals of higher heat content and/or from starting the casting process with heated sand. This latter factor is clearly reflected by the shrinkage values in the alloy of 0.00925 inch per inch and 0.007 inch per inch for 80°F and 160°F silicon sand, respectively. The larger dimensioned cast engine blocks resulting from the.use of heated silica sand simply reflect the larger expansion of the sand mold as a result of the higher sand temperature.
TABLE II
Average Coefficient Sand Sand Shrinkage of Variation Type Temperature (in Lin) of Shrinkage S7.liCd 160F 0.0070 600 SllICa 80F 0.0093 500 Chromite 80F 0.0119 35%
Silicon 80F 0.0110 37%
carbide Spherical 80F 0.0106 360 Carbon sand From the above table, it can be seen.that the use of chromite sand, silicon carbide sand, and carbon sand, resulted in a greater metal shrinkage rate in inches per inch than when using silica sand thus enabling an unbonded sand Core to more Closely follow the shrink-age of the alloy.
Equally important is the fact that the use of Chromite sand, silicon carbide sand and Carbon sand produced a substantially lower coefficient of variation of shrinkage in the metal casting, as compared to the use of silica sand. This means that the shrinkage at the various locations of measurement was more uniform and had less variance than that measured when using silica sand.
Repeat measurements on a cast 250 HP, V-6, 3 liter, marine engine block gave similar results. The ambient temperature, 80°F, silica sand yielded a shrink-age value of 0.0094 inch per inch, and the ambient temperature chromite sand yielded a shrinkage of 0.0118 inch per inch. In addition, the ambient temperature silica sand gave a precision, reflected by the coefficient of variation (of shrinkage), more than 500 less than that obtained through use of the Chromite sand.
These results further evidence that silica sand, as the molding material, produces larger dimensional engine blocks than the use of chromite sand. Moreover, the precision obtained with silica sand is significantly less ~~.~~3~1 than the precision obtained with chromite sand. The test results also indicate that the geometry differences between a V-6 engine block and an in-line three--cylinder block do not materially affect the shrinkage values obtained for the two different sand types.
The particle size of the sand should be in the range of 0.001 inch to 0.015 inch. An expendable pattern casting process requires a relatively coarse sand because of the sand permeability requirements and metal fill time requirements which are interrelated. In expendable pattern casting, the molten metal cannot completely fill the sand mold cavity until the products of foam decomposition have entered the interstices between the sand grains.
Not only are the thermal properties of the sand important in providing precision castings, but it has also been found that the temperature of the sand influences the casting. For example, in winter candi-tions in the foundry, the sand temperature may be in the range of 18.3°C (65°F) to 29.4°C (85°F). In summer, where the ambient temperature may be up to 32.2°C (90°F) or higher, the sand temperature can be in the range of 29.4°C (85°F) to 40.5°C (105°F). With the higher sand temperature in summer, the castings will have a somewhat larger dimension than castings produced in the winter with the sand at a lower temperature. Therefore, to compensate for this differential in dimensions in the cast part, the size of the expendable foam patterns can be adjusted. The dimension of the pattern can be changed by aging the plastic beads before molding, or by aging the molded parts after molding, or by selecting another foam bead type. Thus, by proper aging or selection of the beads, a larger pattern can be obtained which can be used in the winter to compensate for the lower sand temperature, thus resulting in cast parts which have the same dimensions regardless of the temperature of the sand.
Fig. 2 further illustrates the importance of the sand temperature on the precision of casting. Fig. 2 is a curve showing average measurements of an engine block dimension in inches as a function of the tempera-s Lure of unbonded silica sand used in an expendable pattern casting process. The engine block was cast from a hypoeutectic aluminum-silicon alloy having the composition of Example 3 above. As seen in Fig. 2, the average engine block dimension when using sand at ambient temperature of 80°F was 9.53 inches. As the sand temperature was increased to 160°F, the average block dimension also increased. to a value of about 9:59 inches, or an increase of 0.06 inch.
While the above curve shows the difference in dimensions obtained by using silica sand at various tempertures, similar expansion results, although smaller by a factor of approximately four, are obtained using chromite sand, silicon carbide sand, or carbon sand, thus indicating that sand temperature is a factor in obtaining precisely dimensioned castings.
' The invention is based on the discovery that more precise castings can be produced in an expendable pattern casting process by utilizing sand having specific thermal properties and cntrolling the sand temperature or correlating the sand temperature with the pattern size.
When dealing with hypereutectic aluminum silicon alloys, the invention provides a second and unexpected advantage in that the particle size of the precipitated silicon crystals is reduced which substantially improves the 30~ machinability of the alloy.
Various modes of carrying out the invention are contemplated as being within the scope of the following claims particularly pointing out and distinctly claiming the subject matter which is regarded as the invention.
Claims (5)
1. A method of casting a hypereutectic-aluminum silicon alloy, comprising the steps of forming a pattern from an expendable polymeric foam material having a configuration corresponding to the article to be cast, positioning the pattern in spaced relation to an outer flask, introducing an unbonded flowable sand into the flask and surrounding said pattern, said sand having a heat diffusivity greater than 1500J/m2/°K/s1/2=, preparing a molten hypereutectic aluminum-silicon alloy, feeding said molten alloy into contact with the pattern to thereby decompose the pattern with the products of decomposition being entrapped with the interstices of the sand, solidifying the alloy to precipitate particles of primary silicon and produce a cast article, said precipitated particles of primary silicon having an average particle size less than 30 microns, and a coefficiency of variation less than 50%, and removing the cast article from the flask.
2. The method of claim 1, wherein the cast article comprises an engine block for an internal combustion engine.
3. The method of claim 1 or 2, wherein said alloy comprises 12% to 30% by weight of silicon, 0.4% to 5.0% by weight of magnesium, up to 0.3% by weight of manganese, up to 1.45 by weight of iron, up to 5.0% copper, and the balance aluminum.
4. The method of claim 1, 2 or 3, wherein the sand is selected from the group consisting of chromite sand, silicon carbide sand, olivine sand, carbon sand, and mixtures thereof.
5. The method of any one of claims 1 to 4, wherein the sand has a linear expansion from 0°C to 1600°C of less than 1% and a particle size in the range of 0.001 inch to 0.015 inch.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US94048592A | 1992-09-04 | 1992-09-04 | |
| US940,485 | 1992-09-04 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2105361A1 CA2105361A1 (en) | 1994-03-05 |
| CA2105361C true CA2105361C (en) | 2001-01-02 |
Family
ID=25474912
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA 2105361 Expired - Lifetime CA2105361C (en) | 1992-09-04 | 1993-09-01 | Method of expendable pattern casting using sand with specific thermal properties |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA2105361C (en) |
-
1993
- 1993-09-01 CA CA 2105361 patent/CA2105361C/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| CA2105361A1 (en) | 1994-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5000244A (en) | Lost foam casting of dual alloy engine block | |
| US5355930A (en) | Method of expendable pattern casting of hypereutectic aluminum-silicon alloys using sand with specific thermal properties | |
| US5355931A (en) | Method of expendable pattern casting using sand with specific thermal properties | |
| US4966220A (en) | Evaporable foam casting system utilizing a hypereutectic aluminum-silicon alloy | |
| CA1057932A (en) | Method of making an aluminum cylinder heat valve seat coating transplant | |
| Ben Saada et al. | Assessment of the effect of 3D printed sand mold thickness on solidification process of AlSi13 casting alloy | |
| US3701379A (en) | Process of casting utilizing magnesium oxide cores | |
| EP0470968B1 (en) | Shape casting in mouldable media | |
| CA2105361C (en) | Method of expendable pattern casting using sand with specific thermal properties | |
| US6766850B2 (en) | Pressure casting using a supported shell mold | |
| US5612393A (en) | Casting core composition | |
| JP2918180B2 (en) | Use of crushed and graded ore, preferably magnetite ore, for mold and core making | |
| Gopinath et al. | Effect of solidification parameters on the feeding efficiency of Lm6 aluminium alloy casting | |
| JP2549865B2 (en) | Composite sand mold | |
| JPH03501229A (en) | Evaporative foam casting system using hypereutectic aluminum silicon alloy | |
| Yucel et al. | Effects of changing size-weight parameters on the temperature dependent exothermic riser sleeve properties | |
| Omidiji | A Statistical Analysis of Evaporative Pattern Casting Process Parameters for the Production of Aluminum Alloy Components | |
| Yadav et al. | Investigating the Effect of Different Process Parameters on Defects in A713 Aluminium Alloy Castings Produced by Investment Casting Process | |
| Ping | Precision cast dies produced by a ceramic mould process—a review | |
| Fras et al. | Computer modeling of fine graphite eutectic grain formation in the casting central part | |
| Okorafor | Some considerations of the volume shrinkage of aluminium-silicon alloy castings produced in full moulds | |
| GB2285628A (en) | Casting core composition | |
| Wang | Influence of process parameters on microstructure and mechanical properties of lost foam aluminum castings.(Volumes I and II) | |
| Vigh et al. | The Influence of Processing Variables on the Dimensional integrity of Spheroidal-graphite Iron Castings | |
| Er et al. | Mechanical properties and microstructures of AlSi7 alloy cast in sand and in polystyrene-containing sand moulds |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request |