Riserless casting of spheroidal graphite cast iron in igid moulds Dr. F. MAMPAEY, Foundry Research Centre, Grotesteenweg:Noord 2,9710 Zwijnaarde Belgium, Abstract The uence of chemical composition, pouring temperature, nodule count. residual magnesivm content. maul strength and section size ‘changes onthe soundness of spheroidal graphite cast on, poured i chemically bonded sand moulds in the absence Of externa risers, has been Investigated. The major effect ongnates om the casing geomet. As opposed to chunky shapes, a sound plate casting could not be Obtained The presence of section size changes results n casting defects. The amount of porosines comes muni for eutectes commpostons: decreases when lowering the pouring temperature or the residual Magnesium contort, Nadule count has no influence onthe soundness of Unisactional castings. Mechanical properties of unsound spheroxsa graphite cast ion are consterabl less than i sour tera Riassunto Gi autorihanno anaizat influenza dela composiione chim, dela temperatura colata, del numero d nodul del contenuto di magnesio residuo, dels durezza dels forma e dela variaztone dela sozion, su integita del gettin hiss sferodala. colton forme in sab legate ‘himicamenta, massenza cimateroze esteme. (lel principal provocato dala geometra del gett, A ciferenza Ge get di gosso, Spessore. none possibile ottenore un goto forma d pista sottlee sano La presenza dh modiichenellacmensione dela sev0%e db origne a diet nei geo Peruna ghisa a composizone eutettca, la quant i porostisulta essere minima echminuisce con sbbassamento dela temperatura di clata ‘dl contenuto i magnesio residual numero di radu non esercitaalcuna influenza sula santa de gett sezione costante Le propret meecaniche ei gettin ghisa seradale non san sono sgniieatvamente infernal propria iscontrate nel materiale sano. Introduction The solidification of cast iron with eutectic composition is accompanied with a net volume expansion. It results from the expansion during the precipitation of graphite which is more pronounced than the austenitic contraction. The phenomenon is known for quite long (1) and has been studied frequently by dilatation experiments, e.g. (2), (3). The favourable solidification behaviour is taken advantage of to reduce or eliminate the external risering of flake graphite iron castings. Itis surprising that the same beneficial effect does not ‘occur when casting eutectic spheroidal graphite cast iron, Despite the similar quantities of the eutectic phases austenite and graphite in flake and spheroidal {graphite cast iron, the latter often shrinks to the same. extent as steel, Experiments show however, that for a constant mould cavity, the volume change during solidification is independent of the graphite shape (4), 6) Several causes have been put forward to explain the difference in feeding demands of both cast iron types. ‘As opposed to lamellar graphite cast iron, no solid skin is formed during the solidification of spheroidal graphite cast iron, The solidification morphology of the eutectic composition may be classified as mushy (6), (7). The ‘spheroidal graphite precipitation rate differs essentially from the lamellar one since a considerable amount of all ‘graphite precipitates during the initial solidification ‘tages (8). On the contrary, lamellar graphite precipitates progressively during the eutectic solidification, The accompanying volume expansion will be far more intense for the spheroidal graphite spheroidal iron than for the lamellar one which results inan increased pressure on the mould wall during Vol 6(2}(1968) solidification and enlarges the mould cavity (2), (8), 8) ‘The practical consequence of the solidification behaviour of spheroidal graphite cast iron is that without the aid of external risers, sound castings may only be obtained using firmly rigid moulds, The present investigation aims to study some important process variables which might influence the soundness of spheroidal graphite iran castings, poured in chemically bonded sand in the absence of external risers, Experimental Procedure The effect of various factors which might affect the soundness of riserless castings has been studied by pouring two prismatic geometries having the same ‘modulus of 1.5 om: a plate-like type (400 x 400 x 39 mm) anda chunky one {120 x 120 x 60 mm). Both shapes were poured using one mould composed of ‘two moulding boxes dimensioning each 615 x 615 x 125 mm. All moulds have been produced in furan resin bonded sand having a compression strength of 800 N/ cm?. Ina second series of experiments, test castings ‘composed of different sections have been examined (Table 3). All cast iron heats have been melted in a medium frequency induction furnace. Charges have been composed with pig iron, steel scrap and ferro- alloys. The chemical composition of all heats is listed in Table 1. Carbon equivalent values are calculated as Si% + P% CE = c% + SEF Magnesium treatment and inoculation conditions areTABLE | - Chemical composition of all heats (weight percent) NO c Si Ni Mg cE 1 3.59 2.02 0.85 0.089 427 2 3.83 2.16 0.94 0.078 458 3 372 221 0.88 0.057 4.47 4 3.59 2.23 o8t 0.055 434 5 3.68 2.25 0.88 0.067 444 6 331 219 0.93 0.078 4.05 7 3.48 212 0.87 0.067 4.20 8 3.98 2:32 O94 0.073 476 9 379 2.82 ot 0.064 47a 10 3.79 2:30 0.89 0.067 457 1 3.64 216 o.88 0.065 437 2 3.79 2.20 091 0.058 453 13 3.82 223 0.86 0.076 457 14 3.75 2.08 0.93 0.066 4.45 | 15 3.79 219 0.86 0.051 4.53 16 373 20a 0.87 0.063 4.42 7 3.65 2.23 0.90 0.067 4.40 18 371 242 0.89 0.085 4.42 19 3.76 221 0.85 0.066 4.50 20 371 229 O87 0.068 4.48 21 3.84 213 1.68 0.165 4.56 22 374 2.28 0.90 0.073 451 23 3.58 201 0.84 0.069 423 24 374 212 0.92 0.072 4.45 25 3.69 2.02 0.85 0.062 437 26 379 247 084 0.085 4.52 a 3.90 1.99 1.06 0.080 457 Mn = 007% ma P= 0020-- 0.30% S = 0011% max | | TABLE 2 - Magnesium treatment and inocul: ion conditions No Mg-treatment Inoculation Normal conditions 1% 05% A Exceptions, 3 1%. 0.5% B 4 1%. 0.3% A 9 1%. 1.0% B 10 1%. 0.3% A u 1%. 1.0% B 16 1% 1.0% A 20 1%. 1.0% B 21 2%. 05% A Inoculant A = Fe Si (70%) Ca (1%6) Al (4% trade mark VP216) inocuant B = Fe Si (70%) Ca,B;Rare Earth metals fade mark SPHERDX) Metallurgica Science and Technologylisted in Table 2. All castings were top-poured. The soundess has been examined radiographically as wel {as by density measurement, The last technique has been described by Sinha and Kondic (11). Itenables to calculate the amount of porosities by sectioning each casting into 6 (chunky casting} or 12 (plate casting) pieces, The highest density s taken as a reference for porosity calculation of the particular casting. As ‘opposed to the density measurements, radiographic examination only gives a qualitative result. For this reason porosity measurements are most appropriate to quantify the soundness of the ‘experimental castings, Radiographic examination has been used to classify castings as being sound or unsound. Results and discussion Influence of the carbon equivalent The effect of the Carbon Equivalent on the amount of porosities is shown in Figure 1. Variations in Carbon Equivalent are achieved by changing the Carbon content only. ‘When lowering the Carbon Equivalent below the eutectic value, the amount of porosities in both casting shapes drastically increases. Since lower Carbon Equivalents result in decreasing amounts of graphite which precipitate during solidification, the Fig. 1-Variatin a te casting paasitya5 netin othe Caton Equivalent Ip omgevate T3S0C eats 1-11 it aol ieee | catnin Vol 612) (1988) — | Fig. 2- Elect of te prng temperature yan the casing porosity (CE 44-457; nts 2,810.12" 18.171 accompanying volume expansion drops, giving a ‘smaller self-feeding capacity of the iron. ‘Augmentation of the Carbon Equivalent from eutectic to hyper-eutectic compositions slightly raises the porosity percentage of the chunky casting, Influence of the pouring temperature Examination of Figure 2 which represents the casting porosity as a function of the pouring temperature Feveals that a raise in pouring temperature moderately increases the amount of porosities in both test ‘castings. It results from a higher liquid contraction ‘while the eutectic expansion remains the same at constant Carbon Equivalent. Influence of nodule count Changes in nodule count have been realized by varying the amount and the type of inoculant as well as by enlarging the time between the inoculation treatment and the pouring, ‘The presence of carbides has been avoided carefully in all test castings, Within the range of nodule counts realized in the actual experiments, no single correlation with the amount of porosities could be detected for either test castings (Figure 3)Fg 8-Anauntof grossa funtion othe cle count paring tama 13506, CE ASH" A857; pats? 5 10.1,17, 20) | - Nodules / mm Fig, Variation of tho castig pray a a unetinfteresidl magrsir eto (ing temperate 1950; CE 4-46, Pets 23.510 1117.71) TABLE 3 - Influence of section size change: inoculated 1% A) Influence of residual magnesium content Increasing residual magnesium content gradually raises the amount of porosities in both test castings (Figure 4) The effect becomes very pronounced for extremely high residual magnesium percentages. Influence of mould strength ‘Spheroidal graphite iron castings which are produced in the absence of external risering often present internal defects. Experience shows that a reinforcement of the ‘mould may resultin better quality castings. In order to evaluate the influence of the mould strength an the ‘amount of porosities of the test castings, a two dimensional finite element analysis has been carried out. This technique enables to calculate the internal stresses within the mould resulting from the volume expansion occuring during solidification of the casting In the present analysis the thermal influence of the casting on the moulding sand has been neglected. Calculation of o von Mises offers a good method to represent the internal stress distribution within the mould. The maximal value of g von Mises is taken as a measure to compare the mechanical loading of the mould configurations studied. Each of the plate and cchunky casting has been moulded with low (125 mm) and high (205 mm) moulding boxes. The amount of internal porosities within each casting is plotted as @ function of the maximal value of a von Mises obtained {pouring temperature 1350C; Heat No Casting geometry Radiography Porosity % 9 Von Mises (Figure 6) is) max MPa average A sound 0.02 - 0.05 3.93 22 80 ‘unsound (pronounced) 0.22 6.14 23 BI unsound (slightly) O17 6.14 24 B2 sound 012 6.14 26 co unsound (pronounced) 097 5.96 a ct sound O17 5.96 [+) amount of porosities in the massive casting part measuring 120 x 120 x 60 mm. ‘Metalurgical Science and Technology[ — CF emsbima | Fig. 5-CarelatinBetaen the amount of porosities and signa von Mises (pig tumor 1350; CENA, Pats 25 for gh mouing oes: reap aes Torbw rousirg boxes), using the finite element analysis of the moulding box involved (Figure 6). Calculations were executed for an internal pressure at the interface between mould and casting of 1 MPa (9). As aresultof its shape, the chunky casting induces much lower stresses in the ‘mould than the plate casting. Figure 5 shows that the amount of porosities ineary increases with the ‘maximal stress as calculated in the mould. Influence of sec n size changes Most industrial castings are composed of different section sizes. In the present study, these cases have been simulated by adding a prismatic part with smaller section size at the chunky test casting. Dimensions and, shape of this second series of test castings are shown in Figure 6, the experimental results are listed in Table 3, The introduction of a section size change increases the amount of porosities which results in radiographically detectable casting defects (BO, CO) ‘Smoothing the section size transition by rounding the comers, decreases the intensity of the defects. although remaining higher than in the unimodal casting (A). Rounding does not lower the maximal value of 0 von Mises. Consequently, the decrease of the porosities cannot be attributed to a strengthening of the mould but will probably result from changes in the solidification progress. The preceding experimental results show that the chunky shape can be cast radiographically sound within Vo 6/2) (1988) l Fi, Clas set of the casing geometries stuse imerssns nm a catings measure 120 mm parca the cas sects saw a wide range of test conditions (pouring temperature, nodule count, residual magnesium content, carbon equivalent). By reviewing the actual experimental data as listed in Table 3, it seems that the chances to eliminate shrinkage defects in castings presenting section size changes, might be dependent from the ratio between the modulus of the adjacent casting parts, Mechanical properties In order to study to which extent the mechanical properties are lowered by the presence of porosities, Tensile test pieces were machined from the edge {sound location) and the centre (porosity location} of each plate casting. In this way, tensile strength ratio and elongation ratio between sound and unsound material properties are plotted as a function of the porosity percentage of the casting (Figure 7, 8 resp). Examination of Figure 7 reveals thatthe tensile strength in unsound regions equals circa 60 percent of the value obtained for sound spheroidal graphite cast iron. The elongation however drastically drops down to circa 10 percent of its normal value (Figure 8). On the other hand both ratios do not vary with the amount of porosities. This may be explained by the fact that with increasing amounts of porosities, the volume in which the porosities are located, enlarges accordingly rather than gradually lowering the density ofthe restricted spot.fe, 7 -Ratobstwoen esl svorgth in unsound and sone mater as funtion of te porosity percentage of te plate cast Tensile strengtn rato Porosity % Fi. 8-Ratiobetoeen elongation n unsound nd sound materia as ein Ine prosty percentage a esate eastig Conclusions The present study allows to formulate next conclusions: — The geometry of a spheroidal graphite iron casting influences its soundness in a decisive way. As as result of the expansion present during the eutectic, solidification of spheroidal graphite cast iron, particular geometries [in casu plate shape} give raise to stresses in the sand, which the mould cannot withstand. In this case the effect of the ‘mould wall movement exceeds the self-feeding capacity of the spheroidal graphite cast iron, Without the aid of external risering, no sound, castings can be obtained, — The amount of porosities present in spheroidal graphite cast iron castings is influenced by all factors which change the shrinkage behaviour of the iron. Minimum porosities are obtained for low pouring temperature and eutectic compositions, — The soundness of uniformly sectioned castings is not influenced by the nodule count or the inoculation treatment. Increasing the residual magnesium content raises. the amount of porosities. ‘The presence of section size changes results in casting defects which are difficult to avoid. — The mechanical properties of unsound spheroidal raphite cast iron is considerably less than in sound material. Especially, nearly all elongation is lost by the presence of porosities. REFERENCES (1) Schmiat, WA, E. Sulvan and H.F. Taylor, Risering of gray iron tastings, Trans. Am Founar ‘Soc, 62 (1954), 70.77 (2) Nandon, G_and J. Dut Versuchsmethode fur die Kontrolie der Gehuge der GuBeiseniegierungen durch Linge. und Ausden- ‘nungsdruckinderungen wahrend der Kristallisation, ain Iter Foundry Congress Budapest (1978). no. 15 (3) Hummer A, Speisungsbedert und Lingendnderungen wah- rend der Erstarrung von Guseisen Mit Kugolgraphit- Folgerun- ‘gen fr die Speiserbemessung, Giesseret Pras (1985), 241-254 (4) Engler Sand HJ. Woytas, Lunkerverhalten von GuBeisen mit Lameliengrephit und Kugelgraphit sowie die BoeinfluBbarkeit ier Volumenteildefiita Creseorstrschung, 31 (1979) 47-44 (9) Schmicte. Wand 8. Engler, Gestaltinderungen erstarrender GuBstucke in Sandtormen Giossere, 74 (1987) 814.619, (0) Mampacy, &, A quantitative study on the solidification rorp hrology of cast irom. Fondero Beige, (198d), 3-15. (7) Engler S. Zur Morfologie erstarrender Eisen - Kohlenstoft Legiorungen, Giessere! TVIB, 17 (7968) 163-202 (8) Mamosey. F. An experimental study on the feeding beh. viour of cast iron, Fondene Beige, 4 (1805), 3:25 (9) Bako, K, Einige Besinflussungsgr®en bei der Herstellung mafigenauer Guistucke, Giessore! Rundschau, 231976), 1520. (10) Devo. 1... Guny and learcoles, Experiences en vue 'aimentation des pisces moulées en fonte & graphite sphero. dal Forno, 31 1570) S000 (11) Sinha, NP. and V. Konic. Theory and practice of feeding ‘spheroidal graphite iron castings, 21¢h Inter Founcry Congress Bolgum, 12974) no. 7. ‘Metalurgical Science and Technology