LINDHOLM RAGNAR (SE)
WO1992006810A1 | 1992-04-30 | |||
WO1991013176A1 | 1991-09-05 |
1. | A method relating to the production of cast iron castings of the kind which contain graphite crystals in compacted form and entailing controlling and correcting the composition of a cast iron melt, the method comprising preparing a cast iron 5 melt and adding to the melt a structure modifying agent and nucleating agent in amounts according to experience just enough to give fully compacted graphite iron when solidified directly, characterized by extracting a sample of melt in i a) a sample vessel where the inner wall is partially coated with a material which will lower the concentration of dis¬ solved elementary magnesium by at least 0.003% or a corre¬ sponding percentage of some other modifying agent in the melt in the vicinity of that part of the wall, and partially 15 coated with a relatively inert coating not affecting the level of dissolved elementary magnesium in the melt; or in b) two sample vessels where the inner wall of one of the vessels is partially coated with a material, which will lower 20 the concentration of dissolved elementary magnesium by at least 0.003% or a corresponding percentage of some other modifying agent in the melt in the vicinity of that part of the wall, and the other vessel is partially coated with a relatively inert coating not affecting the level of dissolved oό elementary magnesium in the melt; and by using at least three temperature responsive means, at least one of which is arranged in the centre of the vessel/ vessels, at least one of which is placed close to the coating 30 affecting the level of dissolved elementary magnesium, and at least one of which is situated close to the inert coating, and letting said thermal responsive means reach thermal equilibrium with the melt, recording the temperatures regi stred by the temperature responsive means, evaluating from 35 the recorded curve in a manner known per se the characteris¬ tics of the melt, and correcting the inoculating agent and/or nucleating potential and magnesium content or the content of said other modifying agent so that compact graphite crystals are formed during all the pouring and solidification time that is needed in the foundry process. |
2. | A method according to claim 1, characterized in that a sample vessel, where the inner wall is partially coated with a material, which will lower the concentration of dissolved elementary magnesium by at least 0.003% or a corresponding percentage of some other modifying agent in the melt in the vicinity of that part of the wall, and partially coated with an inert coating not affecting the level of dissolved elemen¬ tary magnesium in the melt, is used. |
3. | A method according to claim 1, characterized in that two sample vessels where the inner wall of one of the vessels is partially coated with a material, which will lower the con¬ centration of dissolved elementary magnesium by at least 0.003% or a corresponding percentage of some other modifying agent in the melt in the vicinity of that part of the wall, and the other vessel is partially coated with an inert coa ting not affecting the level of dissolved elementary magne¬ sium in the melt, are used. |
4. | A method according to claim 3, characterized in that the sample veaaels are filled in unison. |
5. | A method according to anyone of claims 1 3, characteri¬ zed in that the coating affecting the level of dissolved elementary magnesium comprises 0 5% sulfur, 0 10% oxides of silicon, manganese or iron and/or 0 0.5% of oxides of potassium and sodium. |
6. | A method according to claim 5, characterized in that the coating affecting the level of dissolved elementary magnesium comprises 3 5% sulfur. |
7. | A method according to anyone of the preceeding claims, characterized in that the temperature responsive means are thermocouples. |
8. | A method according to anyone of claims 1 6, characteri¬ zed in that the temperature responsive means are means for carrying out pyrometrical measurements. |
In the foundry industry it is often desirable to be able to determine in which configuration a certain molten cast iron will solidify. One way of carrying out such determinations is to perform a thermal analysis of the melt. A small sample of the molten metal alloy is taken and is allowed to solidify. During this process, the temperature is measured as a func¬ tion of time. The configuration is then determined by compa¬ ring the obtained cooling curve and its time derivative with reference curves. Such thermal analysis methods are disclosed in e.g. WO86/01755», W091/13176 and WO92/06809.
In the above mentioned methods, a sample of molten metal is obtained by immersing a sample vessel into the molten iron, whereafter said sample is allowed to solidify. The thermal analysis is performed by using temperature responsive means, normally two thermocouples. One of said means is positioned in the centre of the vessel and the other near the vessel wall.
In the method according to W092/06809 a sample vessel where the inner surface is coated with a reactive wall is used. The reactive coating, comprising oxides of silicon, manganese, iron and/or sodium, reacts with active magnesium in the sampled iron and lowers the level of active magnesium in the wall region of the sample vessel.
By using a coated sample vessel according to WO92/06809 it is possible to perform more accurate predictions of solidifica- tion structure compared to the state of the art as represen¬ ted by WO86/01755. In particular, the consumption of magne¬ sium in the near-wall region simulates the natural fading of magnesium during the casting period and provides a predictive warning of magnesium loss. While this feature is indespen-
sible for for the reliable production of compacted graphite iron, it is of great importance to be able to increase the accuracy further.
Summary of the invention
Now it has turned out that by using at least three temperatu¬ re responsive means instead of two and by using a sample vessel where a part of the inner surface is coated with a layer of a substance which will lower the concentration of dissolved elementary magnesium, and the rest of the inner surface is coated by an inert or non-reactive substance (oxides of alumina and zirconia for example) , or the sample vessel itself is non-reactive (quartz or steel) , it is pos- sible to further increase the accuracy of the solidification structure predictions of the molten cast iron. Alternatively, two different sample vessels, one being coated with a reac¬ tive layer and the other one being essentially inert or non- reactive, or coated with an inert or non-reactive substance, may also be used in the present method. Suitable temperature responsive means are thermocouples or pyrometrical means.
A sample of the molten cast iron that is to be analyzed is taken and poured into said sample vessel, which is partially coated with reactive and non-reactive coatings. At least one of the temperature responsive means is placed in the centre of the sample vessel, at least one means close to the inner surface to which a reactive coating has been applied, and at least one means close to the inner surface, to which a non- reactive coating has been applied. Each of said temperature responsive means is then used to record the temperature of the melt during the solidification process as a function of time. The obtained curves are then evaluated in order to determine the configuration of the solidified cast iron sample.
Detailed description of the present invention
The present invention will be explained in more detail with reference to the figures, in which 5
Fig. 1 discloses typical cooling curves that can be obtained by using the method according to WO86/01755;
Fig. 2 illustrates a diagram showing the nodularity percen- 0 tage as a function of the magnesium percentage. In this diagram 0 % nodularity corresponds to a complete compacted graphite cast iron (CGI) , whereas 100 % corresponds to a com¬ plete spheroidal graphite iron (SGI) ; Finally, values below 0 % nodularity relate to grey cast iron. Actually, 0 % nodu- 5 larity corresponds to 100 % compacted graphite cast iron and the bottom of this axis corresponds to 100 % flaky grey cast iron;
Fig. 3 presents cooling curves obtained by the present in- 20 vention; and
Fig. 4 discloses an example of a sample vessel equipped with three thermocouples that can be used in the method according to the present invention.
?5
As previosly mentioned, the procedure for predicting in which configuration a molten cast iron will solidify according to W086/01755 makes use of two thermocouples, one arranged in the centre of the sample vessel and the other near the vessel
30 wall. These two thermocouples provide two distictly different cooling curves which are illustrated in Fig. 1.
The description of key points on these curves is properly made in WO86/01755. A second consideration to be made during 35 the interpretation of the cooling curves is taught in WO92/0- 6809. In this case, a reactive coating on the sample wall consumes magnesium from the melt and causes the near-wall region to solidify as grey iron. The latent heat liberated by the precipitation and growth of grey iron graphite flakes
cause a "deflection" in the normal wall cooling curve. The grey iron/compacted graphite iron (CGI) transition and the "deflected" wall curve are illustrated in Fig. 2 and 3 res¬ pectively.
As already mentioned Fig. 2 shows the amount of spheroidal graphite iron (SGI) and grey cast iron, respectively, as a function of the magnesium level (structural modifyer) . If the melt composition is at point A, the wall reduction of 0.003% Mg as taught in W092/06809 results in the near-wall melt attaining composition B which is still in the CGI plateau and therefore there is no deflection in the wall curve (as in Fig. 1) . However, if the initial melt composition is at point C, the same wall reduction of 0.003% Mg results in the forma- tion of flake iron (Point D) in the near-wall region.
It has turned out that all the measurement points referred to in W086/01755 and WO92/06809 are essential information points. One of the data points referred to in W086/01755 is the minimum temperature on the wall curve prior to eutectic recalescence, so-called T v . However, in the presence of a wall reaction due to flake graphite formation, the minimum temperature on the wall curve is altered or "masked" by the latent heat liberated during flake formation.
Figure 4b illustrates an example of three cooling curves measured simultaneously under controlled conditions, from the same sample of molten compacted graphite iron (CGI) with and without wall reaction. In order to ensure a wall reaction as described in WO92/06809, one sample cup was coated with a 3% sulfur solution while the other sample cup was coated with a less reactive coating. The three cooling curves are described in more detail, according to labelling on Fig. 4b, as fol¬ lows:
Ta 3% Centre thermocouple from sample cup contai¬ ning 3% sulfur wall coating.
Tb 3% Wall thermocouple from sample cup contai¬ ning 3 % sulfur wall coating.
Tb 1% Wall thermocouple from sample cup contai¬ ning 1 % sulfur wall coating.
Figure 3 clearly shows a normal centre thermocouple cooling curve and two different wall cooling curves. The wall thermo¬ couple adjacent to the 3% sulfur coating has a typically obvious wall reaction initiating at the point A (approx. 60 seconds, 1143°C) and continuing to point B (approx. 100 seconds, 114 °C) . The ultimate result of this "shift" in the curve is that the minimum temperature on the wall curve is approximately 2.5°C higher in the presence of a wall reac¬ tion. This is extremely important from a production point-of- view since a difference of 2.5°C in measured undercooling can correspond to the need for +/- 0.5 kg inoculant per tonne to avoid either excessive nodularity (-0.5 kg/tonne) or carbide formation (+0.5 kg/tonne) according to WO86/01755.
The use of three thermoresponsive means in the above mentio¬ ned manner provides an unforeseen advantage. Firstly, the present invention allows two distinctly different types of information to be collected. WO 86/01755 teaches the impor¬ tance of simultaneously controlling the magnesium and ino¬ culant content of the molten iron to remain within the CGI specification window. The present invention not only allows accurate measurement of both values (inoculant efficiency and proximity to grey iron) but it also allows for the coating to be extremely reactive (containing sulfur rather than redu¬ cible oxides as in WO 92/06809) without compromizing the ability to measure the inoculating efficiency. This, for example, when casting components with a long pouring (small castings) or a long solidification time (large castings) allows a highly reactive coating to be used to simulate the Mg-fading, which was previously not possible in the method according to WO 92/06809 due to the adverse affect on the measurement of nucleating potential.
Fig. 4 discloses a sample vessel that can be used in the method according to the invention. The inner wall of a cru¬ cible (8) is partially coated with
I a reactive coating (10) containing 0 - 5% sulfur, 0
- 10% of oxides of silicon, manganese or iron, or 0
- 0.5% of oxides of potassium and sodium; and
II an relatively inert coating (9) such as aluminium or zirconium oxide.
Three thermocouples (1,2,3) arranged in such a way that one (1) is situated in the centre of the crucible (8) , one (2) close to the reactive coating (2) and the last one (3) close to the relatively inert coating (9) , are fixed in a ceramic plug (5) and contained inside a cardboard tube (6) .
Next Patent: METHOD AND APPARATUS FOR SENSING OXYGEN CONCENTRATION