MXPA00004454A - Iron castings with compacted or spheroidal graphite produced by determining coefficients from cooling curves and adjusting the content of structure modifyng agents in the melt - Google Patents

Abstract

The microstructure in which a certain cast iron melt will solidify can be predicted with high accuracy by carrying out four independent calculations and then choosing the calculation giving the best result. The calculations are preferably carried out by a computer.

Description

IRON COATINGS WITH GRAPHITE. COMPACTED OR SPHERICAL PRODUCED BY DETERMINING COEFFICIENTS OF COOLING CURVES AND ADJUSTING THE CONTENT OF X. S AGENTS OF MODIFICATION OF STRUCTURE IN THE CASTING MASS The present invention relates to an improved method for predicting the microstructure with which a certain melt will solidify. cast iron. The invention also relates to an apparatus for carrying out the method.

BACKGROUND OF THE INVENTION _; WO 86/01755 (incorporated by reference) describes a method for producing cast iron with compacted graphite when using thermal analysis. A sample is taken from a batch of cast, molten iron, and this sample is allowed to solidify for 0.5 to 10 minutes.The temperature is recorded simultaneously by two temperature sensitive media, one of which is placed in the center of the sample and the other in the immediate vicinity of the container wall.The so-called cooling curves that represent the temperature of the iron sample as a Time function are recorded for each of the two temperature sensitive media. According to that document then it is possible to determine the necessary amount of structure modification agents that must be added to the melt in order to obtain the desired microstructure. However, detailed information on how to evaluate the curves is not given. * WO 92/06809 (incorporated by reference) describes a specific method for evaluating the cooling curves obtained by the method of WO 86/01755. According to this document, a first plateau in the cooling curve indicates that the flake graphite crystals have precipitated close to the temperature sensitive medium. Since the sample container is intentionally coated with a layer of material that has oxide or sulfide that consumes the active form of the structure modification agents, and thus simulates its loss • natural or fading during the casting period, this plateau can often be found in a cooling curve of a temperature-sensitive medium arranged near the wall of the container. The expert person it can then determine if any structure modification agent has to be added to the melt, in order to obtain cast iron with compacted graphite when using the calibration data. The method of WO 92/06809 requires "perfect" curves comprising a distinct plateau. However, sometimes the cooling curves without a different plateau are recorded despite the fact that squamous graphite has formed. Until now, it has not been possible to use curves without a different plateau as a basis for calculating the precise amount of the structure modifying agent that must be added to the melt in order to produce cast iron with compacted graphite during the entire casting period .

BRIEF DESCRIPTION OF THE INVENTION Now, it has turned out that it is possible to use virtually any set of cooling curves obtained for the eutectic and sub-eutectic solidification, and by the WO equipment. 86/01755 and WO92 / 06809 as a basis for calculating the precise amount of the structure modifying agent that must be added. The method of present invention comprises the steps of: a) determining the amount of the structure modification agent to be added to the melt, in order to obtain the cast iron with compacted graphite, or the cast iron with spheroidal graphite, as a function from where ? - (TAraa¡¡ TAm? N) / (T Bmax T Bml n) and where TAmax is the local maximum value of the cooling curve recorded in the center of the sample container; TAm? N is the local minimum value of the cooling curve recorded in the center of the sample vessel; TBma? is the local maximum value of the cooling curve recorded on the wall of the sample container; TBm? N is the local minimum value of the cooling curve recorded on the wall of the sample container; b) determining the amount of the structure modification agent that is to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite, as a function of f, where f = (TA'max) / (TB'ffiax) where TA'max is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container. c) determining the amount of the structure modifying agent that is to be added to the melt in order to obtain cast iron with compacted graphite, with the cast iron with spheroidal graphite as a function of the area (pB) of the first peak of the first derived from the cooling curve recorded on the wall of the sample container; d) determining the amount of the structure modification agent to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite as a function of K, where K = sA / sE where sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of the sample vessel; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. e) recording the cooling curves in the center of the sample container and in the wall of the sample container, respectively, for a particular sample of cast, cast iron; f) depending on the result in e) choose one of the calibration curves from step a) -d) that gives the most accurate result; and g) calculating the amount of the structure modification agent that is to be added to the melt.

DETAILED DESCRIPTION OF THE INVENTION As already mentioned, the present invention relates to an improved method for predicting microstructure in which a certain cast iron melt will solidify. By using the present method, it is possible to evaluate a much larger range of temperature-time curves compared to the state of the art and it is also possible to obtain more accurate results. The term "cooling curve" as used herein refers to graphs that represent temperature as a function of time, graphs that have been recorded in the manner described in WO86 / 01755 and WO92 / 06809. The term "sample container" as described herein, refers to a small sample container that, when used for thermal analysis, is filled with a sample of molten metal. The temperature of the molten metal is then recorded during solidification in a suitable manner. The walls of the sample container are coated with a material that reduces the amount of structure modification agent in the melt in the immediate vicinity of the container wall. Preferably, the sample container is designed in the manner described in WO86 / 01755, WO92 / 06809, W091 / 13176 (incorporated by reference) and WO96 / 23206 (incorporated by reference). The term "sampling device" as used herein, refers to the device comprising a sample container equipped with at least one temperature-sensitive medium for thermal analysis, this medium being proposed to be immersed in the sample of solidification during the analysis, and a means for filling the sample container with molten metal. The sample container is preferably equipped with sensors in the manner described in WO 96/23206. The term "structure modification agent" as used herein, refers to "compounds that either promote the spheroidization or precipitation of the graphite present in the cast, molten iron." Suitable compounds can be chosen from the group of immaculation substances well known in the art, and form modifying agents, such as magnesium, cerium and other rare earth metals The ratio between the concentration of the structure modification agents in the cast, cast irons and the morphology of the graphite of the cast irons, solidified has been discussed already in the documents cited above in WO92 / 06809 and WO86 / 01755. The invention also relates to an apparatus for controlling the production of cast iron with compacted graphite, apparatus that takes a sample of cast, cast iron, uses the present method to calculate the necessary additions, if any, of the modifying agents of cast iron structure, cast and provides cast iron, cast with the amount of structure modification agent. The apparatus comprises a sampling device, a computer-based data acquisition system, and a means for administering the structure modification agents to cast, cast iron. The sampling device contains a representative sample of cast, cast iron that undergoes thermal analysis during which time / temperature measurements are transmitted to a computer and are presented in the form of cooling curves. The computer calculates the necessary amount of structure modification agent that must be added and automatically activates the means to administer the structure modification agent, so the mass melt is supplied with an appropriate amount of these agents. The invention will now be described with reference to the accompanying figures in which: Figure 1 is a cross section through a part of a sampling device that can be used in conjunction with the present invention; Figure 2 describes examples of cooling curves recorded with two temperature sensitive media, one which is arranged at the midpoint of the sample container (curve I) and the other near the wall of the container (curve II); Figure 3 shows a cooling curve corresponding to curve II in Figure 2. The first time derivative of the curve is also described; Figure 4A defines the parameters TB'max, TBmax, TBm? n. The figure shows the values of TB and crB for the part of a cooling curve of the wall region comprising a growth in steady state and recalescence of subcooling, conventional in the wall region. The central parameters of the curve are marked in general with a capital A while the parameters of the wall are marked with a capital letter B. Figure 4B shows three different appearances of the curve depending on the amount of flake graphite growth during the initial stages of solidification; Figure 5 demonstrates currents in a sample of molten solidification metal and how these currents affect the cast iron layer with flake graphite, typically formed in the vicinity of the vessel wall; Figure 6 is a schematic presentation of an apparatus for controlling the production of cast iron with compacted graphite according to the present invention. As mentioned above, Figure 1 shows the metal-containing portion of sampling device 200 that can be used when carrying out the present method. The means for filling a sample of molten metal in a sample container is not shown. The device 200 is equipped with two sensors, arranged essentially in accordance with the teaching WO 86/01755 cited above. The temperature perception part 210 of the first sensor temperature sensitive 220 is placed in the center of the molten metal 30, and the temperature sensing part 230 of the second sensor 240 is arranged in a location close to the inner surface 60 (which may or may not be coated; sample) of the inner wall 50. A sensor support member 250 is provided to retain the sensors 220, 240 in position during the analysis. The sensor support means is connected to the container by the legs 255, between which molten metal flows into the container when immersed. Figure 2 shows an example of a set of recorded cooling curves of two temperature sensitive media, one that is arranged in the intermediate point of sample container (curve I), and the other near the container wall (curve II). Curve I is a typical curve for the solidification of compacted graphite in the center of the sample. The first point of inflection, or thermal stress, is caused by the formation of primary austenite, which is common for strained, hypoeutectic irons. In contrast, the inflection point in curve II indicates the local formation of flake graphite caused by an insufficiency of the structure modification agent after the reaction with the wall coating. Curve II and its corresponding first time derivative is also described in Figure 3. In this case, there is a relationship between the area of the first peak (pB) of the first time derivative of the cooling curve and the amount of formation of flaked graphite in the vicinity of the wall of the vessel. When the foundry / probe solidifies in a mold / sample container, any oxygen, sulfur, etc., in the atmosphere or in the material of the mold / sample container can react with the structure modification agents in the cast iron. For cast irons with compacted graphite this can result in the formation of flake graphite near the wall of the mold / sample container. In fact, the amount of flake graphite formed is greater when the concentration of the structure modification agents is lowered. Therefore, the amount of flake graphite formed in the wall can be used as a measure of the concentration of structure modification agents, residuals in the volume of the metal. Because the flake graphite nucleates at a higher subcooling temperature than compacted graphite, it can be distinguished by thermal analysis. Figure 3 shows a cooling curve in the corresponding first derivative recorded near the wall where both the flake graphite and the compacted graphite are formed. The amount of flake graphite formation can be monitored by measuring the area pB of the first peak of the first derivative of the temperature-time curve. The amount of compacted graphite formation can be monitored analogously by measuring the area sB of the second peak of the first derivative of the temperature-time curve. However, because the shape of the cooling curve sometimes it is not possible to calculate either or both of the p and s areas defined above. Examples of curves recorded near the wall that are divergent from the ideal curve shape (curve II in Figure 2 and Figure 3) are given in Figure 4B. So far it has not been possible to evaluate results as represented by the TB ?, TB2 and TB3 curves, and in cases where these curves obtained the measurement that is going to repeat, resulting in lost productivity and possibly rejected iron due to excessive temperature loss. According to the present invention, an analysis of the cooling curves can be based on the following fact: As the amount of graphite formation in flakes increases, the amount of graphite formation must decrease • compacted since the total amount of carbon released is approximately constant. Figure 4A shows a cooling curve recorded near the wall in relation to a case where only compacted graphite is formed. The formation of compacted graphite is characterized by the maximum, positive slope of the curve (T'Bmax), recalescence (TBmax - TBmin) and the area sB. Figure 4B exhibits the same curve with progressively increasing amounts of flake graphite formation. Both the recalescence, the maximum slope and the area under the peak of T'B decrease as the amount of graphite in leaflets increases. The amount of heat released by the initial formation of flake graphite in the region near the wall is very small, and in insufficient reality to depend on a control parameter. However, if the bottom shape of the sample container is predominantly spheroidal; and if the container itself is pre-heated (for example by immersion in the molten iron), thus preventing the formation of a solidified iron cooling zone in the region near the wall; and if the container is allowed to hang freely so that the heat is not removed in a floor or mounting position, a favorable convection current will develop within the molten iron contained in the sample container. This convection current "rinses" the flaked graphite away from the preheated upper walls of the sample container and effectively concentrates leaflet growth in a region separated from the flow at the base of the essentially spheroidal container. By strategically placing the wall sensor within the area separated by the flow, a larger and more sensitive measurement of the flake graphite wall reaction is obtained. The method of the present invention requires four calibrations in order to be carried out, specifically: a) determine the amount of the structure modification agent to be added to the melt, in order to obtain the cast iron with compacted graphite, or the cast iron with spheroidal graphite, as a function of?, where Y = (TAn - TA ".) / (TB" ~ TBmin) and where TAmax is the local maximum value of the cooling curve recorded in the center of the sample container; TAm? N is the local minimum value of the cooling curve recorded in the center of the sample vessel; TBmax is the local maximum value of the cooling curve recorded on the wall of the sample container; Bmin is the local minimum value of the cooling curve recorded on the wall of the sample container; b) determine the amount of the structure modification agent that is to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with graphite spheroidal, as a function of f, where f = (TA ':) / (TB' where TA'max is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB'max is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container. c) determining the amount of the structure modifying agent that is to be added to the melt in order to obtain cast iron with compacted graphite, with the cast iron with spheroidal graphite as a function of the area (pB) of the first peak of the first derived from the cooling curve recorded on the wall of the sample container; d) determining the amount of the structure modification agent to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite as a function of K, where K sA / sE where sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of the sample vessel; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. Naturally, the corresponding calibrations are carried out when cast iron is produced with spheroidal graphite. Most calibrations are based on the cooling curves recorded in the center of the sample vessel. The reason for this is that there is usually no flake formation in the center and therefore, TAmax -TAmin, TA'max and sA are not adversely affected by the precipitation of flake graphite. The center can therefore be used as a point of reference even when the modification is low, so that the flake graphite is formed on the wall. The amount of the structure-forming agent that has to be added to a Particular sample is calculated after carrying out a thermal, conventional analysis as described in the previously cited documents, WO 86/01755 and WO92 / 06809. The cooling curves are then analyzed by determining?, F, pB and K. Three independent determinations of the amount of modifying-structure agents that have to be added are carried out, and then it is simple for the person skilled in the art. Choose the determination that gives the most accurate result. It is preferred to carry out the prediction method when using a computer controlled system, especially when a large number of measurements must be carried out. In this case, the same kind of sampling device 22 as described is used. This computer controlled system is summarized in Figure 6. During the measurement of a particular sample, the two temperature sensitive means 10, 12 send signals to a computer 14 comprising a ROM unit 16 and a RAM unit 15 to order to generate the cooling curves. The computer has access to the calibration data mentioned above in the ROM unit 16 and calculates the amount of structure modification agent that must be added to the melt. This amount is signaled to a means 18 for administering the structure modification agent to the melt 20 to be corrected, whereby the melt is supplied with an appropriate amount of these agents.

Claims (12)

CLAIMS 1. A process to produce a cast iron with compacted graphite, a cast iron with spheroidal graphite, which requires a sampling device, a means to monitor the temperature as a function of time and a means to administer structure modification agents to a cast, cast iron, from which casting is to be produced, the method comprising the steps of: a) for the chosen method of function and carrying out the following calibrations: i) determining the quantity of the agent of modification of the structure that is going to be added to the melt, in order to obtain the cast iron with compacted graphite, or the iron cast with spheroidal graphite, as a function of?, where
1. Y = (TAmax - TAm? N) / (T Bma x - T Bm? N) and where TAmax is the local maximum value of the cooling curve recorded in the center of the sample container; min is the local minimum value of the curve of cooling registered in the center of the sample container; TBmax is the local maximum value of the cooling curve recorded on the wall of the sample container; TBm? N is the local minimum value of the cooling curve recorded on the wall of the sample container; ii) determine the amount of the structure modification agent that is to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite, as a function of f, where
2. (TA'max) / (TB ' where TA'max is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB'max is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container. iii) determine the amount of the structure modification agent that will be added to the melt in order to obtain cast iron with compacted graphite, with cast iron with spheroidal graphite as a function of the area (pB) of the first peak of the first derivative of the cooling curve recorded on the wall of the sample container; iv) determining the amount of the structure modifying agent that is to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite as a function of K, where
3. K = sA / sE sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of the sample container; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. b) during solidification record the calibration curves in the center of a sample container and on the wall of the sample container, respectively, for a particular sample of cast, cast iron; c) calculate the control coefficients,?, f, pe and that are related to the temperature-time curves obtained in step b) and choose one of these coefficients?, f, pB and K that gives the most accurate result; d) calculating the amount of structure modification agent (Va) to be added to the melt; e) adding the calculated amount of the structure modification agent; and f) carrying out the casting operation in a manner known per se. "2. A process according to claim 1, characterized in that a sample container is used, essentially spheroidal, and in which the cooling curves recorded near the wall of the container are recorded in a separate area of flow in the base of the sample container, essentially spheroidal 3. A process according to claim 1 or claim 2, characterized in that the cast iron, with compacted graphite, is produced 4. A method for determining the amount of structure modification agent that has to be added to the cast iron, cast in order to produce a cast iron with compacted graphite, or a cast iron with spheroidal graphite, method that requires a sampling device, a means to monitor the temperature with a function of time and a means to administer structure modification agents to a cast, cast iron from which it will be produced casting, the method comprising the steps of: a) for the chosen method of function and carrying out the following calibrations: i) determining the amount of the structure modification agent to be added to the melt, in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite, as a function of?, where
4. ? = (TAmax - TAm? N) / (T Bmax - T Bm? N) and where TAmax is the local maximum value of the cooling curve recorded in the center of the sample container; Amip is the local minimum value of the curve of cooling registered in the center of the sample container; TBmax the maximum local value of the cooling curve recorded on the wall of the sample container; TBmin is the local minimum value of the cooling curve recorded on the wall of the sample container; ii) determine the amount of the structure modification agent that is to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite, as a function of f, where
5. F = (TA'max) / (TB'max) where TA'max is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB'max is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container. iii) determine the amount of the structure modification agent that will be added to the melt in order to obtain cast iron with compacted graphite, with cast iron with spheroidal graphite as a function of the area (pB) of the first peak of the first derivative of the cooling curve recorded on the wall of the sample container; iv) determining the amount of the structure modifying agent that is to be added to the melt in order to obtain cast iron with compacted graphite, or cast iron with spheroidal graphite as a function of K, where
6. K = sA / sE where sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of the sample vessel; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. b) during solidification record the calibration curves in the center of a sample container and on the wall of the sample container, respectively, for a particular sample of cast, cast iron; c) calculate the control coefficients,?, f, pB and K that are related to the temperature-time curves obtained in step b) and choose one of these coefficients?, f, pB and K that gives the most accurate result; d) calculating the amount of the structure modification agent (Va) that has to be added to the melt. A process according to claim 4, characterized in that an essentially spheroidal sample container is used, and in which the cooling curves recorded near the container wall are recorded in a separate flow area at the base of the sample container essentially spheroidal. 6. A method according to claim 4 or claim 5, characterized in that iron casting with compacted graphite is produced. 7. An apparatus for establishing, in real time, a quantity of a structure modification agent that is to be added to a cast iron melt during the process to produce a cast iron with compacted graphite; the apparatus comprises: a first temperature sensor for recording a cooling curve in the center of a sample container; a second temperature sensor for recording a cooling curve in the vicinity of the wall of the sample container; a computer device for determining a quantity value of a structure modification agent that is to be added to the melt; a memory means that is provided with pre-recorded data of the flow curve, the computer device that is adjusted to establish a first control coefficient,?, (from which a first prediction value (VI) can be calculated ) where: ? = (TAmax - TArai n) / (T Bmax - T Bmi n) and where TAmax • is the local maximum value of the cooling curve recorded in the center of the sample container; min is the local minimum value of the curve of cooling registered in the center of the sample container; TBmax the maximum local value of the cooling curve recorded on the wall of the sample container; TBmin is the local minimum value of the cooling curve recorded on the wall of the sample container; the computer device that is adjusted to establish a second control coefficient, f, (from which a second prediction value (V2) can be calculated), where f = (TA'max) / (TB'max) where TA'max is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB'max is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container, the computer device that is adjusted to try to establish a third control coefficient (pB), (from which you can calculate a third prediction value (V3)), where the third control coefficient (pB) refers to the first peak of the first derivative of the cooling curve recorded on the wall of the sample container; the computer device that is adjusted to try to establish a fourth control coefficient (K), (from which a third prediction value (V4) can be established), where
7. K = sA / sE and where sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of the sample container; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. the computer device that adjusts to compare the first, second, third and fourth control coefficients (?, f, pB and K) with the data of the cooling curve, pre-registered, and the computer device that it fits to choose one of the control coefficients (?, f, pB and K) in response to the result of the comparison, and where: the computer device is adjusted to calculate the precise quantity value (Va) of a modifying agent of structure that will be added to the melt in response to the chosen control coefficient (?, f, pB and K). An apparatus according to claim 7, characterized in that the second temperature sensor is arranged in such a way that the cooling curves recorded near the wall of the sample container are recorded in a separate flow area at the base of a essentially spheroidal sample container. 9. An apparatus for establishing, in real time, a quantity of a structure modification agent that is to be added to a cast iron melt during the process of producing an iron cast with spheroidal graphite; the apparatus comprises: a first temperature sensor for recording a cooling curve in the center of a sample container; a second temperature sensor for recording a cooling curve in the vicinity of the wall of the sample container; a computer device for determining a quantity value of a structure modification agent that is to be added to the melt; a memory means that is provided with pre-recorded data of the cooling curve, the computer device that is adjusted to establish a first control coefficient,?, (from which a first prediction value can be calculated (VI )) where :
8. ? = (TA "- TA" / (TBr TBmin) and where TAmax is the local maximum value of the cooling curve recorded in the center of the sample container; TAmin is the local minimum value of the cooling curve registered in the center of the sample container; TBmax is the maximum local value of the cooling curve recorded in the wall of the sample container; TBmin is the local minimum value of the cooling curve recorded on the wall of the sample container; the computer device that is adjusted to establish a second control coefficient, f, (from which a second prediction value (V2) can be calculated), where
9. (TA '/ (TB'max) where TA'ma? is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB'max is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container. the computer device that is adjusted to try to establish a third control coefficient (pB) < • (from which a third prediction value (V3) can be calculated), where the third control coefficient (pB) refers to the first peak of the first derivative of the cooling curve recorded on the container wall shows; the computer device that is adjusted to try to establish a fourth control coefficient (K), (from which a third prediction value (V4) can be established), where
10. K sA / sE and where sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of 1 sample vessel; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. the computer device that is adjusted to compare the first, second, third and fourth control coefficient (?, f, pB and K) with the data of the cooling curve, pre-recorded, and the computer device that adjusts to choose one of the control coefficients (?, f, pB and K) in response to the result of the comparison, and where: the computer device adjusts to calculate the precise quantity value (Va) of a structure modification agent that is to be added to the melt in response to the chosen control coefficient (?, f, pB and K). An apparatus according to claim 9, characterized in that the second temperature sensor is arranged in such a way that the cooling curves recorded near the wall of the sample container are recorded in a separate flow area at the base of a essentially spheroidal sample container.
11. 11. An apparatus for carrying out the process of claims 1 or 2, the apparatus comprising: a sampling device for taking a sample of cast iron, cast from a cast iron melt from which it is going to produce a foundry comprising CGI or SGI; a first temperature sensor for recording a cooling curve of the center of a sample container; a second temperature sensor for recording a cooling curve in the vicinity of the wall of the sample container; a computer device for determining a quantity value of a structure modification agent to be added to the melt, a memory medium that is provided with pre-recorded data of the cooling curve, a means for delivering a correct amount of an agent of structure modification in response to a signal from the computer device, the signal corresponding to the quantity value; the computer device that is adjusted to establish a first control coefficient,?, (from which a first prediction value (VI) can be calculated), where (TAp 1 ttiin) / TB "TBm? N) and where TAmax is the local maximum value of the cooling curve recorded in the center of the sample container; Amm is the local minimum value of the cooling curve recorded in the center of the sample container; TBmax is the local maximum value of the curve of cooling recorded on the wall of the sample container; Bmin is the local minimum value of the cooling curve recorded on the wall of the sample container; the computer device that is adjusted to establish a second control coefficient, f, (from which a second prediction value (V2) can be calculated), where F = (TA'max) / (TB'max) where TA'max is the maximum value of the first derivative of the cooling curve recorded in the center of the sample container; and TB'max is the maximum value of the first derivative of the cooling curve recorded on the wall of the sample container. the computer device that is adjusted to try to establish a third control coefficient (pB), (from which a third prediction value (V3) can be calculated), where the third control coefficient (pB) refers to the first peak of the first derivative of the curve of cooling recorded on the sample container wall; the computer device that is adjusted to try to establish a fourth control coefficient (K), (from which a third prediction value (V4) can be established), where K = sA / sE and where sA is the area under the second peak of the second derivative of the cooling curve recorded in the center of the sample container; and sB is the area under the second peak of the first derivative of the cooling curve recorded in the container wall. the computer device that is adjusted to compare the first, second, third and fourth control coefficient (?, f, pB and K) with the data of the cooling curve, pre-recorded, and the computer device that adjusts to choose one of the control coefficients (?, f, pB and K) in response to the result of the comparison, and where: the computer device is adjusted to calculate the precise quantity value (Va) of a structure modification agent that is to be added to the melt in response to the chosen control coefficient (?, f, pB and K). the computer that adjusts to send a signal that corresponds to the amount value to the medium, whereby a correct amount of structure modification agent is added to the melt.
12. 12. An apparatus according to claim 11, characterized in that the second temperature sensor is arranged in such a way that the cooling curves registered within the wall of the sample container are recorded in a separate flow area at the base of a essentially spheroidal sample container.