WO2011159234A1

WO2011159234A1 - Method for determining amounts of inoculant to be added to a cast-iron melt

Info

Publication number: WO2011159234A1
Application number: PCT/SE2011/050732
Authority: WO
Inventors: Martin Wennerstein; Tobias BJÖRKLIND; Fredrik Wilberfors
Original assignee: Scania Cv Ab
Priority date: 2010-06-16
Filing date: 2011-06-14
Publication date: 2011-12-22
Also published as: RU2013101763A; RU2528569C2; CN102985811A; SE1050616A1; KR20130043160A; KR101412165B1; BR112012030956A2; EP2583089A4; SE534912C2; EP2583089A1; CN102985811B

Abstract

Method for determining amounts of inoculant to be added to a cast-iron melt in a specific casting process, comprising the steps of: -providing a first sample holder(1) and a second sample holder(2) each having a thermocouple(3, 4) which is connected to analysis equipment (5); -filling each sample holder(1, 2) with an amount of molten iron; recording a first cooling curve during solidification of the iron in the first sample holder(1) and a second cooling curve during solidification of the iron in the second sample holder(2); characterised in that one of the sample holders, before being filled with molten iron, has a predetermined amount of inoculant placed in it which represents the inoculant saturation level in the specific casting process, the amount of inoculant to be added to the melt in the specific casting process being determined on the basis of the difference between the lowest eutectic temperature(TE_low) on the first cooling curve and the lowest eutectic temperature(TE_low) on the second cooling curve.

Description

Method for determining amounts of inoculant to be added to a cast-iron melt TECHNICAL FIELD

The present invention relates to a method for determining the amount of inoculant to be added to a cast-iron melt according to the preamble of claim 1 . BACKGROUND

Cast iron is a commonly used construction material for components of trucks. It is for example used in engine blocks, cylinder heads, cylinder liners, main- bearing caps and spring mountings. As most cast components undergo no subsequent heat treatment or plastic machining to modify their microstructure or remedy defects in them, their characteristics are largely already determined at the casting stage. The solidification structure, i.e. the amount and appearance of the primary phase (austenite) and the eutectic structure (austenite and graphite), is therefore extremely important for the characteristics of the cast item. The amounts of these phases/structures can be determined by so-called thermal analysis based on cooling curves recorded during solidification of a sample from the melt. Thermal analysis is described in, for example, specifications

SE516136, SE515026 and WO97355184.

A known way of using thermal analysis to check phases and structures in cast iron is based on the liquidus temperature of the molten iron, i.e. the temperature at which the melt begins to solidify. Depending on how much the measured liquidus temperature differs from a predetermined value, the structure for the finished cast item can be modified, e.g. by adding carbon. However, this known method does not provide an exact measure of how the internal structures nucleate or grow, but merely an approximate estimate of the respective amounts of the primary phase and the eutectic phase. Another known practice is to control the nucleation of the internal structures of the iron by adding inoculant, but with the known methods it has been found difficult to optimise the amount of inoculant to be added.

The nucleation of the internal structures of the iron therefore greatly affects its characteristics, inter alia its incidence of defects and its strength. For the final characteristics of the finished cast item, the structure of the eutectic phase is particularly important. However, the fact that previously known methods provide no good measure of the structure of this phase can lead to strength problems and scrapping.

The object of the invention is therefore to indicate a method which reliably determines the amount of inoculant to be added to a cast-iron melt and which solves the above problems or at least reduces them to a minimum. SUMMARY OF THE INVENTION

This object is achieved by the method for determining amounts of inoculant to be added to a cast-iron melt in a specific casting process, comprising the steps of:

- providing a first sample holder and second sample holder which each have a thermocouple connected to analysis equipment;

- filling each sample holder with an amount of molten iron; recording a first cooling curve during solidification of the iron in the first sample holder, and a second cooling curve during solidification of the iron in the second sample holder;

characterised in that one of the sample holders, before being filled with molten iron, has a predetermined amount of inoculant placed in it which represents the saturation level for inoculant in the specific casting process, the amount of inoculant to be added to the melt in the specific casting process being determined on the basis of the difference between the lowest eutectic temperature (TE|_0W) on the first cooling curve and the lowest eutectic temperature (TE|_0W) on the second cooling curve. The method provides precise control of the casting process, resulting in only slight variations in the quality of the finished cast item, with consequent savings in terms of less scrap during casting, machining and assembly, and less risk of accidents. According to an alternative, the predetermined amount of inoculant for the casting process is an average value based on saturation levels for inoculant in a plurality of cast-iron melts in the specific casting process.

According to an alternative, the predetermined amount of inoculant for the casting process is a selected value from saturation levels for inoculant in a plurality of cast-iron melts in the specific casting process.

In cases where the inoculant contains silicon, the effect of the latter on the eutectic temperature is with advantage calculated away from the inoculated sample on the basis of a predetermined relationship.

The amount of inoculant to be added in the specific casting process may if so desired be directed towards over-inoculation or under-inoculation of the melt. The method preferably relates to determining the structure of a lamellar graphite type of iron. LIST OF DRAWINGS

Figure 1 Part of an Fe-C diagram

Figure 2 Cooling pattern of a cast-iron melt

Figure 3 Test set-up for conducting an experiment according to the

method of the invention

Figure 4: Diagram showing the difference between the respective eutectic temperatures TE|_OW of inoculated and uninoculated samples from a cast-iron melt. DESCRIPTION OF THE INVENTION

We describe by way of introduction the theoretical background to the invention.

Solidification of a melt involves the initiation of small solid nuclei of

agglomerated atoms, about which nuclei the melt will thereafter solidify. The formation of these nuclei, i.e. the nucleation, is controlled by the total energy in the melt. A system, e.g. a metal melt, endeavours to have as low an energy content as possible. Whether it is the solid state or the liquid state of a substance which has the lower energy content depends on its temperature. The formation of a solid phase in a melt entails the formation of a volume in which the atoms are arranged in an energy-economising way. At the same time, a surface forms between the solid and liquid phases. The atoms in the surface become pressed into positions which they would not normally occupy, and energy is required for them to arrive there. When a melt solidifies to a solid body, there is therefore a decrease in free energy per unit volume G_v and an increase in surface energy γ, provided that it takes place at below the solidification temperature T_M.

Changes in the total energy content of the system may be described by the equation AG = AG_vV + yA , in which AG is the total change in energy, V the volume of the body and A the surface area of the body. Nucleation only takes place if it leads to the total energy content of the system decreasing, i.e. to ΔΘ becoming negative. As the surface energy counters the phase conversion, the melt will remain liquid even when the temperature passes its solidification point. This is called supercooling of the melt. The more the temperature drops below the solidification point, the greater the driving force for phase change. Nucleation takes place when the temperature drops so much that the decrease in energy per unit volume becomes greater than the amount of energy required to create the surface. For homogeneous melts, this supercooling may amount to several hundreds of degrees, but is normally significantly less. How much supercooling is required for nucleation to take place is an indicator of how easily nucleation can take place in the melt, also known as the melt's nucleation potential.

The solidification pattern of a cast iron may be described by an Fe-C phase diagram, as in Figure 1 , with temperature on the vertical axis and carbon content percentage by weight on the horizontal axis. Figure 1 indicates the carbon content in the region which is relevant to cast iron, i.e. up to 5%. The lines in the diagram demarcate various phases which the iron assumes at different temperatures and carbon contents. The diagram features the Iiquidus line 1 and a line which represents the eutectic temperature 2. These two lines are important for predicting the structure of the cast product. At the Iiquidus temperature, austenite, also called γ iron, precipitates. At the eutectic temperature, carbon also begins to precipitate out from the remaining melt. The carbon precipitates in the form of graphite which, depending on its configuration, greatly influences the characteristics of the material.

During cooling from fully molten state to solidified state, a melt passes through the various phases in the Fe-C diagram. Figure 2 depicts

schematically a solidification curve for a cast iron. The plateau "a" at 1200°C denotes the Iiquidus temperature of the melt. At this point, austenite begins to form in the melt. When the temperature passes 1 150°C, the eutectic solidification begins at "b", which is the melt's lowest eutectic temperature TEiow The eutectic solidification is indicated by a slight temperature increase in the melt. When the temperature reverts to steady decrease, region "c", the whole melt has changed to solid form.

How much supercooling is needed for the eutectic precipitation to begin, i.e. TEiow, has been found to be a good indicator of the melt's nucleation potential. Little supercooling means good nucleation potential. It is important to achieve high nucleation potential during the eutectic solidification, since this results in even precipitation of graphite during solidification. Even precipitation of graphite is necessary for good mechanical characteristics of the cast item, since the graphite's volume increase during solidification counteracts the austenite's volume decrease.

The nucleation potential may be controlled by adding nucleation points in the form of inoculant. Adding inoculant raises the temperature at which the eutectic solidification takes place. In other words, less supercooling of the melt is required for the eutectic solidification to begin.

It is important to optimise the content of inoculant added to the melt. Adding too little inoculant may cause uneven or insufficient graphite precipitation, leading to formation of carbide and, in the worst case, to so-called white solidification. Adding too much inoculant leads to high production costs and may also result in shortcomings in the structure of the cast product, e.g. due to graphite expansion.

As mentioned above, it is therefore advantageous to determine the melt's nucleation potential at the eutectic minimum, point "b" on the cooling curve. This is because the supercooling during the eutectic solidification has a direct bearing on the final structure of the cast item. The inventors have found by measurements that a cast-iron melt can become "saturated" with inoculant. This means that during addition of inoculant the temperature for the eutectic solidification stops increasing at a certain level even if more inoculant is added. Consequently, this "saturation level" represents an absolute limit for the melt's TE|_0W, which is the highest eutectic temperature, i.e. the smallest supercooling, which can in practice be achieved in the melt.

For the method according to the invention, described in more detail below, this relationship is important, since it results in a fixed level for how high the eutectic temperature in a cast-iron melt can in practice become. The nucleation potential in an uninoculated melt can therefore be compared with a stable reference value. The method according to the invention is described in detail below.

Step 1

As a first step, a cast-iron melt is made, which is done by melting initial material, e.g. scrap, returns, pig iron and machining waste. The composition and temperature of the melt are monitored and any necessary adjustments are made.

Step 2

As a second step, a thermal analysis test rig is provided. The test rig (see Figure 3) comprises two sample cups 1 and 2, e.g. sand cups, each provided with a thermocouple 3 and 4 respectively. The thermocouples are connected to a set of analysis equipment 5 on which thermal analysis software is run. A predetermined amount of inoculant which represents the inoculant saturation level of the specific casting process is placed in the one sample cup.

The inoculant saturation level, i.e. the amount of inoculant which has to be added to cast-iron melts for them to become saturated with it, varies with the process conditions during the making of the melts. These conditions comprise for example the composition of the initial material (scrap) and the settings of the process equipment. In industrial processes for making cast-iron components, so-called casting processes, however, endeavours are made to keep the process conditions as constant as possible, i.e. to keep them within specific limits. In making certain components, e.g. engine blocks, endeavours are always made to use scrap with a composition which is kept within stated limits and/or always to use the same casting equipment for making a given component. This is done so that as the results of different casting operations may be as uniform as possible.

The amount of inoculant to be added in the one sample cup to ensure that the melt placed in it will reach saturation level may therefore be determined beforehand as a representative value for a specific casting process. The representative value may thereafter be used each time a new cast-iron melt is to be made in the casting process. This representative value may for example be a mean value of inoculant saturation levels determined for a plurality of melts for a specific casting process, e.g. a process for casting engine blocks of a certain type. The representative value may also be selected from a plurality of different inoculant saturation levels determined on a plurality of melts for a specific casting process. For example, the representative value may be a median value or a maximum value or a minimum. The representative value for the saturation level for the specific casting process is then stored, e.g. in a nonvolatile memory in a computer, from which it can be retrieved and used each time the specific casting process is to take place. It is also possible to determine inoculant saturation values for a plurality of different cast-iron melts and thereafter adapt a linear relationship to these saturation values. The way the amount of inoculant required for an individual cast-iron melt to become saturated is determined is as follows. The eutectic temperature TEiow is determined simultaneously for two samples taken from the melt. The one sample is uninoculated and the other has inoculant added. For saturation level determination, a sample from the melt is poured

simultaneously into each of two identical test cups, followed by determination of TEiow for each sample from cooling curves recorded during solidification of the samples. The nucleation potential of the inoculated sample is then determined from the difference between TE|_0W of the inoculated sample and TEiow of the uninoculated sample. The procedure is repeated several times and the one sample is inoculated with progressively more inoculant each time than in the previous batch. By comparing the nucleation potentials of the repeated samples it is possible to identify the level at which the nucleation potential ceases to increase with increasing inoculant content. This level is the respective melt's inoculant saturation level. As mentioned above, the saturation level is determined for a plurality of melts, on the basis of which an average saturation level for a specific casting process can be determined or a representative value be selected.

Figure 4, discussed in more detail later on, shows how the eutectic temperature TE|_0W stops increasing when more than 0.20 wt% of inoculant is added to a certain cast-iron melt.

The inoculant saturation level is thus determined by repeated experiments each comprising two simultaneous measurements in two separate identical test cups under the same conditions. This is because in thermal analysis whereby cooling curves are recorded during solidification of cast iron in test cups the curves are affected by various factors which are at variance with the nucleation potential of the melt. The following are example of these factors:

• Test cup material

• Variations in thermocouple positioning and configuration

· Variations in air draught due to walls, ventilation etc.

• Amount of melt in cup

• Chemical composition of melt

• Temperature of melt when poured into cup The above distorting factors affect the measurement result by, for example, shifting the whole cooling curve upwards or sideways, thereby placing the eutectic temperature at an incorrect level.

If errors occur during measurements conducted in two test cups in identical conditions, however, both of their measurements will be affected by the same error. These distorting factors are eliminated by subtracting the respective samples' eutectic temperatures TE|_0W from one another.

A further factor affecting the result is the silicon content of the inoculant. The inoculant usually contains silicon, e.g. in the form of FeSi. Silicon alters the chemical composition of the melt in the inoculated test piece. As the temperature at which the eutectic solidification begins is based on the chemical composition of the melt, the eutectic temperature in the inoculated sample increases when silicon is added. To exclude also this distorting factor, the effect of silicon on the eutectic temperature is subtracted from the inoculated sample.

It is also possible to arrive at a relationship for the influence of silicon on the eutectic temperature. This may be done by examination of phase diagrams and by linear adaptation. It is also possible to arrive at the relationship by a suitable calculation programme, e.g. ThermoCalc. The relationship can be used to calculate away the influence of silicon on the eutectic temperature.

The inoculant may also be alloyed not with silicon but with other substances which affect the equilibrium lines in the phase diagram, i.e. they affect the eutectic temperature. It is also possible to arrive at relationships for how these substances affect the eutectic temperature, and to use these relationships to compensate the eutectic temperature in the inoculated sample.

Step 3

As a third step, the test cups are then filled with melt, and cooling curves are recorded during solidification of the samples. From the two cooling curves, the lowest eutectic temperature TE|_0W for each sample is then determined. The lowest eutectic temperature of the uninoculated sample is then subtracted from the lowest eutectic temperature of the inoculated sample.

According to the invention, the effect of any silicon content of the inoculant on the eutectic temperature of the inoculated sample is eliminated before that temperature is subtracted from the eutectic temperature of the uninoculated sample.

Step 4

As a fourth step, the calculated difference in °C is used as a basis for determining how much inoculant has to be added to the melt to ensure that the finished cast item acquires a desired structure. The resulting difference in °C, i.e. the nucleation potential of the melt, serves as a measure of how far the eutectic temperature of the melt is from the practically achievable eutectic temperature. The resulting difference in °C can be related to the final structure of the cast item. The relationship between the temperature difference, the amount of inoculant to be added and the final structure of the cast item can be determined empirically. This is done by making several melts over a substantial period of time, determining by the method according to the invention the amount of inoculant to be added and monitoring the structure of the cast item. For each melt made, the relationship between the number of degrees of temperature difference and the amount of inoculant to be added is adjusted until there is a good enough match with the final structure of the cast item. It is also possible to direct the amount of inoculant to be added to the melt towards over-inoculation or under-inoculation if this is desirable for a certain product.

It should be noted that it is not possible to add to an individual melt precisely the amount of inoculant which corresponds to the saturation level for the specific casting process for the individual melt. This is because the saturation levels of the individual melts over time have a spread about the

representative value for the saturation level of the casting process. For example, the spread may be caused by variation in the composition of the individual melts within the stated limits. This means that the saturation level for inoculant for each individual melt differs from the saturation level for the casting process. Adding inoculant to the individual melts in precisely the amount indicated by the saturation level for the casting process would lead to their being over-inoculated or under-inoculated. As previously mentioned, over-inoculation and under-inoculation have various disadvantages with regard to costs and material characteristics.

Step 5

As a fifth step, the desired amount of inoculant is added to the melt, followed by the melt being poured into a suitable casting mould. It may of course happen that only one melt is to be made and that there is no predetermined value for its inoculant saturation level. In that case, the saturation level is determined by repeated experiments with two test cups as described above. The amount of inoculant determined is added to the melt, which is then used for casting.

DESCRIPTION OF EMBODIMENTS

The invention is described below with reference to a concrete example. A 10 tonne melt of the grey iron type with lamellar graphite was prepared in a medium-frequency furnace from the manufacturer ABB.

As a first step, the amount of inoculant needed for the melt to be saturated with inoculant was determined. To this end, four test batches were prepared, numbered 1 to 4. The sampling equipment, see Figure 3, for each batch comprised two sample holders 1 and 2 in the form of sand cups each provided with a thermocouple 3, 4.

The specifications of the sand cups and thermocouples were as follows:

· Cup material: 65 gram shell-moulded sand cups made of silica

sand.

• Sand coated with 3.5% phenol resin, iron oxide and stearin-based lubricant.

• Thermocouples: NiCr-NiAI alloy insulated by high-purity quartz

glass tubing.

• Thermocouple locations as Figure 3. During the measurements, the average volume of the cups was 4.86 cm³ (based on spot weight of 350 g and density of 7.22 g/cm³). Each thermocouple was connected to a separate channel of the analysis equipment comprising a computer whose processor was used to run ATAS evaluation software from the manufacturer Novacast. The mean weight of the melt accommodated in the test cups was calculated, and inoculant 6 of the Superseed type, an alloy of FeSi with strontium content, was placed in test cup 2 of batches 2 to 4. The amount of inoculant was calculated with respect to the amount of melt accommodated in the test cups, in wt% of cast iron. The amounts of inoculant added appear in Table 1 . No inoculant was put into any of the batch 1 test cups.

The test cups were then each filled with the same amount of melt, and cooling curves were recorded during solidification of the samples and were analysed in the evaluation programme.

The lowest eutectic temperature TE|_0W was then determined for each cooling curve by means of the evaluation programme. This is called TE|_0W

"measured". As the inoculant contained silicon, the effect of silicon on the eutectic temperature was then subtracted from each of the three inoculated samples. The thus corrected samples are called TE|_0W "corrected". Table 1 shows the eutectic temperatures TE|_0W "measured" for samples 1 and 2 in each batch, and the eutectic temperatures for the inoculated samples after compensation for silicon (TE|_0W "corrected"). The eutectic temperature for each inoculated sample TE|_0W "measured" was then subtracted from the eutectic temperature for the respective uninoculated sample. The result is called Δ TE|_0W "measured". After the eutectic

temperatures of the inoculated samples had been corrected for silicon content, they were subtracted from the eutectic temperatures of the

uninoculated samples. The result is called Δ TE|_0W "corrected". The values appear in Table 1 . Test Inoculant Uninoculated Inoculated Inoculated ΔΤΕ,™ ΔΤΕ,™ batch wt% TE|_0W TE|_0W TE|_0W "measured" "corrected"

"measured" "corrected"

1 0 1 138.24 1 138.04 1 138.04 -0.2 -0.2

2 0.1 1 135.975 1 139.725 1 138.975 3.75 3

3 0.2 1 132.425 1 139.075 1 137.575 6.65 5.15

4 0.4 1 132.075 1 139.95 1 136.95 7.875 4.875

Table : Results from determination of inoculant saturation amounts

Figure 4 is a diagram showing ATE|_0W "measured" and ATE|_0W "corrected" for each batch.

It may be seen from the diagram that when the samples have been corrected for the influence of silicon on the eutectic temperature the difference between the eutectic temperatures of the inoculated and uninoculated samples stops increasing when the inoculant content exceeds 0.2 wt%. This content thus represents the melt's inoculant saturation level. The diagram in Figure 4 also shows that when the samples are not compensated for the influence of

silicon, the difference between the eutectic temperatures continues to

increase with increasing inoculant content. An inoculant saturation level can therefore not be determined.

Claims

A method for determining amounts of inoculant to be added to a cast-iron melt in a specific casting process, comprising the steps of:

- providing a first sample holder (1 ) and a second sample holder (2) each having a thermocouple (3,

4) which is connected to analysis equipment

(5);

- filling each sample holder (1 , 2) with an amount of molten iron;

recording a first cooling curve during solidification of the iron in the first sample holder (1 ) and a second cooling curve during

solidification of the iron in the second sample holder (2);

characterised

in that one of the sample holders, before being filled with molten iron, has a predetermined amount of inoculant placed in it which

represents the inoculant saturation level of the specific casting process, the amount of inoculant to be added to the melt in the specific casting process being determined on the basis of the difference between the lowest eutectic temperature (TE|_0W) on the first cooling curve and the lowest eutectic temperature (TE|_0W) on the second cooling curve.

The method according to claim 1 , in which the predetermined amount of inoculant for the casting process is an average value based on inoculant saturation levels of a plurality of melts in the specific casting process.

The method according to claim 1 , in which the predetermined amount of inoculant for the casting process is a selected value from inoculant saturation levels of a plurality of melts in the specific casting process.

The method according to any one of claims 1 - 3, in which the inoculant contains silicon and the effect of silicon on the eutectic temperature (TE|_0W) is calculated away on the basis of a

predetermined relationship between silicon and eutectic temperature.

The method according to any one of claims 1 - 4, in which the amount of inoculant to be added to the melt in the specific casting process is directed towards over-inoculation or under-inoculation of the melt.

6. The method according to any one of claims 1 - 5, in which the iron is of lamellar graphite type.