Influence of Mould Design on the Solidification and ...

Influence of Mould Design on the Solidification and

Soundness of Heavy Forging Ingots*

By Koichi TASHIRO,** Shiro WATANABE,*** Ikujiro KITAGAWA*** and Itaru TAMURA***

Synopsis

Influence of hot top and mould design on the formation of central porosi-ties and loose structure in heavy forging ingot was analysed by using finite element method. The results of the analysis were compared with those of sectioning investigation of 100 and 135 t ingots and the influence of mould and hot top design on the internal defects has been made clear quantitatively.

The result shows that the geometry of hot top and mould design plays most important role in the manufacture of sound heavy ingots. The central

porosities and loose structure are liable to increase when the rate of vertical solidification at the centerline of ingot exceeds the value of about 10 mm/ min, and the defects are strengthened at the area where the rate of solidifica-tion is accelerated.

For 0.25%C-3.5%Ni-Cr-Mo-V steel ingot, "A" segregation be-

gins to form when the rate of transverse (horizontal) solidification de-creases to the value of about 0.8 mm/min.

I. Introduction

Accompanying to the increase in the size of forgings used widely in ships, machines for iron making, and

power generation plants among present-day heavy industries, the ingot size is also becoming larger, each weighing as much as from 400 to 500 t. This increase in ingot size is resulting in a greater size of cavities referred to as loose structures and center porosities which are formed in the center line zone of ingots. These weak and brittle structures in ingots can be reduced to some extent by forging effects brought to bear on the forging process. However, since these forging effects decrease with an increase in ingot size under limited press capacity, it is necessary to com-

pensate for this by increasing the internal soundness of the ingot itself. The soundness of a large forging ingot made by the conventional process depends primarily on the design conditions of mould and hot top. Since there is no suitable way for making a nondestructive examination of the soundness of a manufactured ingot, quality assessment must usually depend on a sectioning in-vestigation of an ingot or of the forged materials. The former, however, involves considerable cost and time, and the latter does not provide an immediate evaluation of ingot soundness owing to forging side effects. Thus, in order to clarify the effects of design conditions of hot top and mould on solidification characteristics of large ingots, the authors prepared a solidification analysis program using a finite element method (F.E.M.) and analysed the solidification char-acteristics under various conditions. Furthermore, the results were compared with the measured results of the mould temperature and of sectioning investiga-

tion of large ingots, and are considered from various

points of view in this paper. The design conditions for reducing loose structures and segregations were

based on these results. Some clues for the design of ultra-heavy ingots weighing 400 t or 500 t could be

obtained.

II, Analytical Procedure

1. Calculation Method

For convenience sake, the finite element method has been used rather than the finite difference meth-od for the computer program calculation when the boundary conditions are complicated.' The calcula-tion of ingot solidification in this paper was also

performed by F.E.M. for the same reason. The shape of an ingot cross section and mould is cor-rugated but this has been regarded as axial sym-metry. An example of element division is shown in Fig. 1. Elements whose temperature changes rapidly were divided as finely as possible to determine the temperature with the greatest possible accuracy.

2. Initial Conditions, Boundary Conditions and Material Properties

A 0.39 % C steel was assumed for the calculations which were carried out with the liquidus temperature

Fig. 1. Element division and temperature measuring points

for 65 t ingot.

*

**

Originally published in Tetsu-to-Hagane, 67 (1981), 103, in Japanese. English version received July 19, 1982. © 1983 ISIJ

Japan Casting & Forging Corporation, Oaza Nakabaru, Tobata-ku, Kitakyushu 804. Research & Development Laboratory, Japan Casting & Forging Corporation, Oaza Nakabaru, Tobata-ku, Kitakyushu 804.

(312) Research Article

Transactions ISIJ, Vol. 23, 1983 (313)

taken as 1 498 °C and the solidus temperature as 1 438 °C. The initial conditions and boundary con-ditions were the same as those in Fig. 2. Emissivity of the outside surfaces of the mould and hot top frame and that of the ingot surface following air gap forma-tion was assumed to be a function of temperature3~ and estimated to be 0.75 0.95. The coefficient of the convection heat transfer on the outside surfaces of the mould and hot top frame was estimated to be 15 kcalJm2•h.°C.

The conditions of each boundary of the hot top frame have been reported by various authors and the boundary conditions in this paper are as shown in Table 1. The boundary condition of the upper sur-face of hot top has been reported as the whole surface being either insulated4~ or heated by an outer source5~ : In this analysis, one part of the surface was assumed to be insulated and the other part maintained at a con-stant temperature as shown in Fig. 2 (1 480 °C be-tween the liquidus and solidus temperatures). Since the hot top of a large forging ingot dealt with in this analysis is high, the boundary condition of the upper surface of hot top does not affect very much the solidification of the ingot body. Changes in the shape of the hot top resulting from shrinkage should also be considered in the case where the hot top is low and the boundary condition of the upper surface affects the solidification; this would be evident from the analysis results.

The properties of carbon steel were used in making the calculations. The properties of cast iron and brick also used in making the calculations are as fol-lows :

(1) Thermal conductivity and specific heat: These properties of steel, cast iron and brick are shown in Fig. 3.

(2) Density: Shown in Fig. 4. (3) Latent heat: That of steel was taken as 60

cal/g. The following methods are known to distribute the latent heat throughout the solidification range7) : (a) A method for equally distributing latent heat in

the solidification range.

(b) A method for distributing latent heat in the form of an isosceles triangle.

(c) A method for distributing latent heat in proportion to the solid fraction.

The method (a) has been used in this paper.

(4) Convection : There are many unknown fac-tors concerning the motion of molten steel after it has been poured. The convection caused by pouring may be considered negligible since the temperature of mol-ten steel at the end of the pouring approximates the liquidus temperature in the case of the considered large ingots in this paper. Consequently, convection was not taken into consideration. Instead, a method using a tenfold thermal conductivity of molten steel was proposed.7j

(5) Other: Acceleration of vertical solidification has been reported for carbon steel; this is considered as a piling up of sedimental crystals but this has not been taken into consideration when making the cal-culations.

III. Experimental Procedure

1. Temperature Measurement Procedure

In order to check the accuracy of the temperature calculated by F.E.M., measurement was made of the

Fig . 2. Initial and boundary

tion analysis.

condition used for solidifica-

Table 1. Boundary condition about hot top.

Fig. 3. Thermal conductivity

solidification analysis.

and specific heat used for

Fig. 4, Density used for solidification analysis.

Research Article

(314) Transactions ISIJ, Vol. 23, 1983

temperature distribution on the conventional hot top frame and ingot mould. Figure 5 shows the points at which temperature measurement was made and the various kinds of ther-mocouples used for the hot top frame and ingot mould of a 65 t ingot (0.39 % C steel). Temperature mea-surement was carried out at 9 points on the chamotte brick of the hot top lining, at 3 points on the cast iron of the hot top frame and at 3 points on the ingot mould. Pt-Pt. 13%Rh sheathed thermocouples (~b 1.6 mm) were used for points l-.6 where the tem-

perature was expected to exceed 1 000 °C and alumel-chromel sheathed thermocouples (~!i 3.2 mm) were used at the other points. The thermocouples used for the hot top lining brick were laterally inserted through vent holes in the cast iron frame.

2. Procedure for the Sectioning Investigation of Ingots

In order to make a comparison of the solidification analysis results with the internal structures of ingot, a sectioning investigation was carried out using 110 and 135 t ingots of 3.5%Ni-Cr-Mo-V steel. Vertical sectioning of ingots was carried out so as to include the center line of the ingots and the transverse sec-tioning was at a distance 10, 50 and 80 % down from the ingot shoulder. Each section was investigated by means of a dye penetrant test, magnetic particle in-spection, microstructure observation, sulphur print, macro-etching, segregation of chemical composition,

porosities and inclusions. The distribution of porosities responsible for the soundness and internal quality of ingots was inves-tigated by measurement of their area ratios and num-ber and size using polished specimens cut from each

position on the center line of the ingots and each

Temperature measuring points

1, 2, 3 : 10Ms(SheathedPR 1.655e) from brick lining surface

4, 5, 6 : 45sis( 1, PR1.6sm) ii

7, 8, 9 : 135( 'I CA3.2sm) ii

10, ii, 12:30ss( ,, CA32550) from outer face of frame

17, 18,19:30ms( I, CA3.2s ) from outer ;face of mould

radial position at a distance 10, 50 and 80 % down from the ingot shoulder. The measurements of 490 fields (on a ~5 25.4 mm vertical cross section) were carried out at a magnification of x 100 with an opti-cal microscope. The area ratio of porosities was counted by the measuring method used for nonmetal-lic inclusions specified in JIS and the number of these in each field was summed up. The breadth and length determination of porosities were made in the same way as nonmetallic inclusions and an evaluation of the results was carried out on the basis of the length values. The range of "A" segregation formation was measured by vertical and transverse section sulphur

prints.

Iv. Comparison of Calculated Results with Measured Results

In order to check the calculated results, the tem-perature of the hot top frame and the ingot mould was measured to see if the calculated results agreed with the measured results. Figure 6 shows a comparison of the calculated with the measured temperatures of the hot top frame and mould of a 65 t ingot. This comparison was per-formed using the temperatures of representative points, that is, the mid points of the height of the hot top frame 2, 5, 8, 10 and the mould 18 (cf Fig. 5). The calculated temperatures of the points 2, 5, 8 of the bricks of the hot top frame were either some-what higher than the measured temperatures or in good agreement with them. The calculated tempera-tures for the point 10 of the cast iron of the hot top frame and the point 18 of the mould were somewhat lower than the measured values. Since the tempera-ture gradient of the hot top frame and mould is steep, the large variability of the temperature is caused by the deviation of the measurement position from the expected position during the setting of the thermo-couple or by the thermal deformation. In considera-tion of this, it may be said that the calculated values are in good agreement with the measured values. The temperature of each position had the following

Fig. 5. Position of temperature measurement and kinds of

thermocouple used for 65 t ingot.

Fig. 6. Comparison of the calculated temp. with the mea-

sured in the hot top frame and mould of 65 t ingot.

Research Article


characteristics : The temperatures for positions 2, 5 of the inside of the brick of the hot top frame were over 1 000 °C but those for position 8 of the outside of the brick and for position 10 of the cast iron of the hot top frame were below 600 °C, which was not very high. The highest temperature corresponded to the inside position and the temperature decreased with a shift to the outside. The temperature for position 18 of the surface re-gion of the mould was also less than 600 °C, and that for position 17 was nearly the same as that of posi-tion 18. The temperature for position 19 was some-what lower than that for position 18.2) The explana-tion for this is that the cooling from the base affects the temperature of position 19 but hardly exerts any influence on that of position 18 (the mid-point of the height of the mould). Since the calculated and measured values agreed as

mentioned above, the properties of steel and the cal-culation conditions assumed are considered reason-able. Subsequent calculations were carried out in the same way to analyse the relationship between mould design and solidification characteristics of steel ingots. Furthermore, the relationship between these characteristics and the internal soundness of large ingots was studied by a sectioning investigation so as to gain an understanding of a method for producing large steel ingots of high quality.

V. Analysis Results and the Discussion

1. Investigation of Solid cation Patterns of Large Ingots

The following figures show the calculation results for the solidification of large ingots, using an XY

plotter. These results describe with the lapse of time the positions at 1 460 °C where the fraction solid is about 60 % in accordance with the proportion of the temperature between the liquidus line and solidus line temperatures (1498 and 1 438 °C, respectively) of the carbon steel whose carbon content is assumed to be 0.39 %. The positions of the 60 % fraction solid are considered to average a substantial solidification front since the solidification front measured by a bar test is not the position for the solidus temperature but for the fraction solid of 60N 70 %.14) Also the position for the 60 % fraction solid is situated almost in the centre of the mushy zone. Hereafter, this line of the 60 % fraction solid is referred to as the solidification line. Figure 7 shows the shape and size of the ingot, which must be taken into consideration for the fol-lowing. 1. Influence of Thermal Conditions of the Hot Top Frame Figures 8 to 10 show the solidification patterns for a 65 t ingot under various thermal conditions of hot top frame. Figure 8 shows the solidification pattern of the ingot with a conventional hot top frame which is the same as that used for the temperature measure-ment in this experiment. Figure 9 shows the pattern in the case in which the insulation of the hot top bricks has become stronger. That is, insulating bricks (isolite) are used for the out-side lining. The dashed line in Fig. 9 shows the

solidification line in Fig. 8 in which chamotte bricks are used as the outside lining. Even in such a case, the solidification line of the hot top is delayed only somewhat and there is little difference in solidifica-tion lines of the ingot body. It is apparent that the effect of insulating bricks on reducing loose structures does not amount to very much. The thermal con-ductivities k of chamotte brick and insulating brick are about 3.0 X 10-3 and 1.0 x 10 3 cal/cm. s. °C re-spectively, the densities p are 1.9 and 0.88 g/cm3 and the specific heats are nearly equal. Consequently, the thermal diffusivity a(=k/cp) of insulating brick is 72 % that of chamotte brick. It follows that insulat-ing effects increase considerably but an increase in insulating effects has no appreciable influence on the solidification pattern of an ingot body. Figure 10 shows the solidification pattern at the time the ;outer surface of hot top frame has been completely insulated. The dashed lines show the solidification lines inyFig. 8 for the outer surface not insulated but there is little difference in these lines. The probable reason for this is that heat in hot top is removed endothermically to the bricks and cast iron of the hot top frame because of their large heat capac-ity, rather than by emission from the outer surface of the hot top frame.8} In order to investigate the effect of a reduction in the heat capacity of the hot top frame, solidification patterns were calculated for the case in which flange in lower part of hot top cast iron frame for supporting bricks was cut down to a smaller size and the bricks and cast iron of the hot top frame have been preheated at 100 °C. The results for these were compared with those obtained by the conven-tional design as shown in Fig. 8. But improvement

Fig. 7. Definition of ingot size.

Research Article


of the heat capacity of the hot top frame did not bring about any significant changes.

As mentioned above, a certain amount of insula-tion or decrease in the heat capacity of hot top frame itself does not bring about improvement in the solidi-fication characteristics of the ingot body. Thus, the shape of the hot top and ingot body will exhibit more important effect. 2. Influence of HID of an Ingot, Ratio of the Hot Top

Diameter to the Ingot Diameter ~ and Ingot Taper Figure 11 shows the solidification pattern for a 135 t ingot used in the sectioning investigation. Fig-ure 12 shows the solidification pattern for a 110 t ingot

(hereafter referred to as the 110 t S-type ingot be-cause of the large cp) whose ratio cP of the hot top diameter to the ingot diameter (cf : Fig. 7) was nearly equal to that of a 135 t ingot and whose height/ diameter ratio H/D was different from that of a 135 t ingot. The cp values of the former and latter are about 87 % and 91 %, respectively, and those for H/D, about 1.4 and 1.1, respectively.

When comparing the 135 t ingot with the 110 t S-type ingot, the unsolidified molten pool of the for-mer is somewhat narrower and deeper than the latter. The former is considered to have the disadvantage for an abatement in the number of central defects. This was confirmed also by the sectioning investigation mentioned below. In order to investigate the influence ofthe diameter

of the hot top on the solidification characteristics, the

solidification pattern of the 110 t ingot used for the sectioning investigation is shown in Fig. 13. This ingot is hereafter referred to as the 110 t L-type ingot since the hot top with a small diameter is used. ~o of this ingot was found to be about 71 % and the unsolidified molten pool became quite narrow and deep within l8'20 hr after pouring, as compared with the 110 t S-type ingot. Consequently, the form-ed dendrite caused bridging and it is considered that feeding of molten steel into the ingot body must surely be obstructed. This was confirmed by the sectioning investigation. Since the 110 t S-type ingot using a hot top frame large in diameter as shown in Fig. 12 has a large ratio of hot top diameter to ingot diameter, it follows that the shape of the unsolidified molten pool should be wider with a shallower V shape than that of the 110 t L-type ingot. The feeding into the lower part of the dendritic structure was not ob-structed; this is considered to prevent the central cavities from occurring.

The influence of the ingot taper on the solidifica-tion characteristics was investigated by the same cal-culation method. With a large taper, there is the natural advantage that the formation of loose struc-tures are prevented. A too steep taper in thickness causes the ingot shoulder to tear (cracks in the lateral direction resulting from the ingot weight below the

point where sliding of ingot body, due to solidification shrinkage, is locally blocked by the mould). Con-sequently, about a 10 % taper is considered reason-

Fig. 8. Solidification pattern of

(H/D=1.3, TF =11.6 %,

65 t ingot. cp=84.9 %) Fig. 9.

Solidification pattern of 65 t ingot with

insulating brick as hot top lining.

Fig. 10.

Solidification pattern of 65 t ingot

insulated at outer wall of hot top.

Research Article

Transactions ISIS', Val. 23, 1983 (317

able for practical design. It was found from the analysis of the results on

solidification patterns under various conditions men-tioned above that the internal soundness of ingots is more affected by geometrical design conditions such as the ratio cp and H/D of the ingot than by thermal conditions coming from design of the hot top frame. Therefore in the following the results of solidification analysis were investigated by a comparison with the sectioning investigation results for the 110 and 135 t ingots.

2. Comparison of Solid jication Characteristics with Section- ing Investigation Results

135 t and 110 t L-type ingots were sectioned so as to investigate the distribution of internal porosities, loose structure morphology and the range of the "A" segregation formation. In the following, the mor-

phology of internal defects in these ingots is inves-tigated and a comparison is made with the solidifica-tion characteristics mentioned above, 1. Relationship between Morphology of Defects in the Core of an Ingot and Solidification Characteristics Photograph 1 shows a sulphur print of the vertical and transverse sections of a 135 t ingot of 3.5 % Ni-Cr-Mo--V steel. Porosities (small cavities) of about 1 mm in maximum size were observed in a part of the core called the V zone and on the segregation lines of the "A" segregation zone of this ingot. But

in Photo. 2, which shows the penetrant test results on a vertical section of a 110 t L-type ingot made of the same kind of steel, shrinkage cavities with cracks, so-called loose structures were easily observed in the core zone and were surrounded by porosities. Figure 14 shows a comparison of the vertical solidi-fication rate with the range and distribution of internal defects along the center line of the 135 t, 110 t S-type and 110 t L-type ingots. The vertical solidification rate in this figure was obtained by the solidification

patterns in Figs. 11 to 13. The conditions for ob-taining sound ingots were investigated qualitatively by the configuration of the solidification pattern, par-ticularly the configuration of the V shaped solidifica-tion front near the center line of an ingot. These conditions could be quantitatively investigated by reference to the vertical solidification rate along the center line as a parameter representing the effects of this configuration. That is to say, as can be seen from Fig. 14, the maximum vertical solidification rate of a 110 t S-type ingot has the smallest value of about 6 mm/min. The 110 t L-type ingot has the largest value of about 15 mm/min and the 135 t ingot, a medium value of about 10 mm/min.

Large loose structures such as those in the 110 t L-type ingot were not formed in the 135 t ingot but small cavities not exceeding 1.0 mm in diameter, the so-called center porosities could be observed in the center line zone as mentioned above. On comparing

Fig. 11. Solidification pattern of 135 t ingot.

(gyp 87 %, H/D :=1.4).

Fig. 12. Solidification pattern of 110 t S-type ingot.

(~ Y 91 %, HJD ; ' 1.1)

Fig. 13. Solidification pattern of ingot. (~o7l %, H/D

110 t L-type l.l)

Research Article


the vertical solidification rate with the distribution of

prosities of the 135 t ingot, it was found that the range where the area ratio of porosities was large, that is, where porosities were formed in large quantities, was

the one at which the vertical solidification rate in-creased or, in other words, accelerated. The position

where large porosities formed was within the range where the vertical solidification rate was large.

When considering this along with the configura-

tion of the solidification front near the center line of ingot whose solidification pattern is shown above, the

following mechanism can be proposed. As seen from the solidification pattern in Fig. 11, since the con-

figuration of the solidification front is U-shaped and the feeder effect is satisfactory in the initial period of

solidification for about 15 hr after pouring when cool-ing from the bottom is effective, porosities hardly

form. However, since the cooling effect from the bottom

decreases rapidly and vertical solidification proceeds

while being strongly affected by lateral solidification in an intermediate period of solidification, the vertical

solidification rate is remarkably accelerated as shown

Photo. 1. Sulphur print of 135 t ingot of 3.5%Ni- Cr-Mo-V steel.

Photo. 2. Result of penetration test 110 t L-type ingot of 3.5%N

on vertical section of

i-Cr-Mo-V steel.

Fig. 14. Comparison of vertical solidifcation

110 t L-type and 135 t ingots, the

ture in 110 t L-type ingot and the

porosity in 135 t ingot.

rate of 110 t S-type,

range of loose struc-

distribution of micro-

Research Article


in Fig. 14. The configuration of the solidification front in this range is sharp and deeply V-shaped as shown in Fig. 11. Thus many porosities form in this range. The probable reason for this is that as the V-shaped configuration becomes sharper, the inter-sections of the dendritic structures get more com-plicated and the feeding of molten steel into the den-dritic space becomes more difficult. The formation of porosities in the accelerated range of solidification is also related to the length of range where feeding is difficult.

As solidification proceeds close to the hot top, the configuration of the solidification front again takes on a wide angle V shape, making feeding easy and con-sequently the formation of porosities decreases remark-ably.

The maximum vertical solidification rate in an 110 t L-type ingot is the highest among three ingots and its position is located close to the hot top as can be seen from Fig. 14. This is thought to be caused by the smaller diameter of the hot top on this ingot by reference to the solidification pattern in Fig. 13. The solidification front particularly 18N20 hr after

pouring is nearly perpendicular to the ingot bottom. Thus, feeding is interrupted by the intersection of dendritic crystals, thus causing bridging. This is con-sidered to be the major cause for the formation of loose structures. The range of loose structure forma-tion is also within the accelerated zone of solidifica-tion as seen from Fig. 14 and bridging is assumed to occur near the maximum accelerated zone. Also, cracks as loose structures were observed even in the zone close to the ingot bottom where the vertical solidification rate was low. However, no such micro-

porosities surround these cracks as observed in ordinary loose structures and feeding effect is considered to be satisfactory according to the solidification pattern in Fig. 13. Therefore, these cracks are considered to originate from those resulting from thermal stress after solidification and from the propagation of cracks during gas cutting. These cracks occurred on the boundary face (grain boundary) of equiaxed crystals which is brittle and weak. But since the H/ D ratio of 110 t S-type ingot is small and the ratio of the hot top diameter to ingot diameter is large, the maximum vertical solidifica-tion rate is considerably low as a result of the cooling effect from the bottom and feeding effect from the hot top. Thus, this ingot has the advantage of taking on sound quality at its core. 2. Approximate Equation for the Maximum Vertical Solidification Rate

As seen from the results above, the value of the maximum vertical solidification rate can be taken as an indication of the formation of loose structures or central porosities; a solidification rate value under 10 mm/mm n is considered to be ideal for about 100 t ingots according to the sectioning investigation of 135 t and 110 t L-type ingots. Since the larger ingots increase the intersection of dendritic crystals in the center line zone in the intermediate period, a lower value of the solidification rate seems necessary to avoid

loose structures. If computer calculations are to be performed to

obtain the maximum vertical solidification rate for each ingot having its own shape, size and conditions of ingot making, too much time and cost are involved. But, if the maximum vertical solidification rate SN (the vertical solidification rate of the maximum point of each curve in Fig. 14) determined by numerical calculation can be described by an approximate equa-tion, it will be convenient. Since the solidification rate is remarkably affected by the ratio of the hot top diameter to the ingot diameter, the H/ D ratio of the ingot and taper as mentioned above, SN can be described in the following approximate equation ST as a function of these factors by the statistical analysis.

ST _ 2 exp (0.586i-0.0558T) 11.72 (HID)

-3 .74ln(co-70.0)+5.99} .....................(1)

where, D:

HJD

M. Kawai pro

does not have t

Mean diameter of ingot (cf. Fig. 7) Ratio of the hot top diameter to the ingot diameter (%) Height/Diameter ratio of the ingot Taper of the ingot (%).

p osed the following equation SK which he term T"P.loi

26'4 0.035 S K = D 1(H/D-O.7)+ 0.69 .........(2) Figure 15 shows a comparison of the numerical cal-

culation values SN for maximum vertical solidification rates along with the approximate values ST. These

are in good agreement. ST has the term TP and the

equation is somewhat complicated, but closer to SN than SK. The approximation degree of SK for SN is

lower than ST but the calculation process is simpler.

3. Relationship between the Transverse Solidification Rate and the "A" Segregation

The sulphur print of the 135 t ingot in Photo. 1 is not very clear owing to the low sulphur content of

0.008 but the formation range of the "A" segregation can be seen from this print which also shows that the "A" segregation forms even at a point near the center

line of the ingot.

Figure 16 shows a comparison of the transverse

Fig. 15. Comparison of approximate

merical analysis value SN of

solidification rate.

value ST with nu-

maximum vertical

Research Article


solidification rate at the mid-height (50 % position) of each ingot body along with the "A" segregation formation range and the distribution of porosities in the 135 t ingot. The vertical solidification rates of the 110 t L-type, 110 t S-type and l35 t ingots are different but their transverse solidification rates at the 50 % position are nearly equal. The "A" segregation formation range for the 135 t ingot is the zone in which the transverse solidification rate is less than about 0.8 mm/min. The U.C.L.

(Upper Critical Limit of solidification rate [m/hr] ) for the formation of the "A" segregation calculated from Eq. (3) presented in the previous paper11 is about 1.1 mm/min owing to the carbon content of 0.25 % in this ingot. This value approximately agrees with that mentioned above.

U.C.L. = 0.174 (% C)+0.025 ............(3)

According to the experiments carried out on a 14 kg ingot (0.64-0.76 % C, 0.0130.015 % S) by K. Suzuki and T. Miyamoto,l2~ the range of the "A" segregation formation is where the relationship be-tween the solidification rate R (mm/min) and cooling rate E (°C/min) is E<8.75R-1.1. However, the "A" segregation for the 135 t ingot investigated in this ex-periment formed within the range where both R and s were remarkably small as shown in Fig. 17. This is probably because the chemical composition, par-ticularly the carbon content is different.

The following can be understood in the light of the relationship between the transverse solidification rate and distribution of porosities in the transverse direc-tion of the ingot in Fig. 16. Since the solidification rate near the ingot surface is high but the feeding is sufficient, the size and number of porosities are small. The porosities become progressively larger but the area ratio increases only a little in the range where the transverse solidification rate starts decreasing and the "A" segregation forms. This is because the range of porosity formation is almost restricted in the "A" segregation lines. However, in the center line zone of the ingot where the transverse solidification rate accelerates, porosities increase and suddenly become larger in a manner similar to the vertical solidification

mentioned above.

VI. Conclusion

The solidification characteristics of large ingots used for forgings were analysed by the finite element method and compared with the sectioning investiga-tion results in order to determine the design condi-tions of the hot top and mould that would reduce loose structures in the core zone of an ingot. The following results were obtained.

(1) In order to make large high quality ingots, the geometric design of the mould and hot top is im-

portant. (2) Loose structures and centre porosities in the

centre line zone tend to form when the maximum solidification rate is higher and to appear especially in the accelerated solidification zone.

(3) The maximum solidification rate can be for-mulated approximately as a function of the mean diameter of an ingot, the height/diameter ratio of an ingot, the ratio of the hot top diameter to the ingot diameter and the taper.

(4) In order to reduce loose structures, it is neces-sary to have design condition under which the maxi-mum solidification rate is less than 10 mm/min for about 100 t ingot. The lower solidification rate is

preferable for larger ingots. (5) The range of the "A" segregation formation

in 3.5 %Ni-Cr-Mo-V steel ingot with 0.25 % C is within the internal zone where the transverse solidifi-cation rate is less than 0.8 mm/min.

As above-mentioned, this paper has quantitatively clarified under what conditions loose structures and the "A" segregation are formed in large ingots used for forgings and how they can be reduced. Through application of the results presented in this paper, it is possible to manufacture heavy high quality ingots of the 400 t class.

Acknowledgements

The authors would like to express their apprecia-tion to the late Dr. M. Kawai, formerly Managing Director of the Japan Casting and Forging Corp., for his guidance and useful comments throughout this study. We pray that his soul may rest in peace.

Fig. 16. Comparison of transverse solidification rate at the

middle part of three ingots. Range of "A" segre-

gation formation and distribution of porosity in

135 t ingot.

Fig. 17. Relationship between the rates of local solidifica-

tion and local cooling, and the position at which "A" segregation formed in 135 t ingot

.

Research Article


REFERENCES

1) 0. C. Zienkiewicz and Y. K. Cheung: The Finite Element Method in Structural and Continuum Mechanics, McGraw Hill Book Co., New York, (1967).

2) K. Tashiro, S. Watanabe and I. Tamura: Tetsu-to-Hagane, 62 (1976), 29.

3) R.J. Sarjant and M. R. Slack: JISI, 177 (1954), 428. 4) S. Maekawa, K. Suzuki and T. Miyamoto : Japan Steel

Works Technical Review, 29 (1971), 73. 5) K. Narita and T. Mori : The 19th Committee (Solidifica-

tion), Japan Society for the Promotion of Science (JSPS), Rep. No. 8691, 34 (May, 1968). 6) G. A. Fischer: Proc. ASTM, 62, 1137. 7) A. I. Koler, J. D. Thomas and A. A. Jzauaras: Cast

Metals. Res. J., 9 (1973), 4, 156.

8) 9)

10)

11)

12)

13)

14)

G. Fentor: JISI,186 (1957), 396. H. Sakuda, H. Ooi and J. Matsuno : Proceedings of the 7th International Forging Conference, Chambre Syndicale de la Grosse Forge Francaise, Paris, (1975), 19. M. Kawai : A private message, Comment on Ingot Mould Design, (1975). K. Tashiro, T. Todoroki and S. Kimura : Tetsu-to-Hagane, 57 (1971), 1479. K. Suzuki and T. Miyamoto: Tetsu-to-Hagane, 63 (1977), 45 & 53. K. Kinoshita, H. Kitaoka, S. Okano and T. Emi : Tetsu-to-Hagane, 65 (1979), 1868. T. Takahashi and M. Kudo: Tetsu-to-Hagane, 56 (1970), 1511.

Influence of Mould Design on the Solidification and ...

Documents

Transcript of Influence of Mould Design on the Solidification and ...