JPH06258129A - Method for measuring molten metal surface level under slag - Google Patents

Abstract

PURPOSE:To exactly detect the molten metal surface level even through a slag layer suspended in a large quantity of conductive metal droplets by supplying alternating current to an eddy current sensor, generating eddy current m a molten metal surface and calculating it on the basis of a generated induction magnetic field. CONSTITUTION:When high frequency current is supplied to an oscillation coil 11 in the neighborhood of a molten metal surface 14, eddy current 15 is allowed to flow in the molten metal surface 14. Thereby an induction magnetic field 16 is produced and a line of magnetic force 17 is detected with receiving coils 12, 13. The strength of the detected line of the magnetic force 17 is fluctuated according to the distance between an eddy current sensor 10 and the molten metal surface 14. Therefore the position of the molten metal surface 14 can be detected from the strength of the line of the magnetic force detected with the coil 12, 13. Next, when measurement is performed as high frequency alternating current of 0.5-500kHz is supplied to the oscillation coil 11, a molten metal surface level can be highly precisely detected without any influence caused by a conductive slag layer 18. The measurement of the molten metal surface, level using alternating current of specified frequency can be performed up to 50vol.% of a dispersion quantity of metal liquid droplets of the slag layer 18 suspended in the molten metal surface 14.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for highly accurately measuring the level of a metal bath covered with a slag layer in which a large amount of metal particles are suspended.

[0002]

2. Description of the Related Art When decarburizing and blowing alloy steel such as stainless steel in a converter, a vacuum degassing device, etc., when carbon in molten steel reacts with blowing oxygen and becomes CO gas and is removed from molten steel. At the same time, some of useful components such as Cr, Fe and Mn are oxidized according to the following reaction. 4 [Cr] + 3O ₂ → 2 (Cr ₂ O ₃ ) 2 [Fe] + O ₂ → 2 (FeO) 2 [Mn] + O ₂ → 2 (MnO)

The metal elements such as Cr, Fe and Mn that have become oxides migrate to the slag floating on the surface of the molten steel. The metal element in the slag is reduced from the oxide to a metal state in the final stage of steel making, and is recovered as molten metal in the molten steel. The recovery is performed by utilizing that metal elements such as Cr, Fe and Mn are easily reduced to a metal state by Si, and for example, a predetermined amount of Si is added to the molten steel in the ladle during vacuum refining. . The reduction reaction with Si is as follows. 2 (Cr ₂ O ₃ ) + 3Si → 4 [Cr] +3 (SiO
₂ ) 2 (FeO) + Si → 2 [Fe] + (SiO ₂ ) 2 (MnO) + Si → 2 [Mn] + (SiO ₂ ) Also, Al, Ti, etc., which have a greater oxygen affinity than Si, are used as reducing agents. At this time, Si in the slag can be transferred to metal.

Cr, Fe, Mn, etc. in the metallic state are taken into the molten steel and the composition of the molten steel is adjusted. In order to adjust the composition with high accuracy, it is necessary to quantitatively grasp the metal elements that migrate from the slag to the molten steel at the final stage. Recently, steel grades with extremely strict standards for Si content have begun to be used. In order to cope with such high-accuracy component adjustment and steel grades in which the Si content is strictly controlled, the amount of the easily reducible oxide reduced by Si contained in the slag is accurately grasped. It is necessary. The amount of easily reducible metal element contained in the slag is C
It is calculated from the oxygen concentration and the amount of slag in metal oxides such as r ₂ O ₃ , FeO and MnO ₂ . The calculated oxygen content is
It serves as a standard for determining the amount of Si required as a reducing agent. As a method of quantifying oxygen in a metal oxide, a method of performing fluorescent X-ray analysis on a slag sample is known. Fluorescent X
In the line analysis, molten slag collected from a converter during refining, a vacuum degassing device, etc. is prepared as an analysis sample by a glass bead method, a press molding method, or the like.

In the glass bead method, as shown in FIG. 1, for example, the solidified slag is crushed and then weighed, 0.2 g of slag is put in a platinum crucible together with 2.0 g of a flux such as sodium carbonate, and a bead sampler is used. Heat and stir to melt and cool uniformly. According to this method, it takes about 25 minutes to obtain a sample for analysis. In the press molding method, an appropriate amount of the collected slag is filled in an aluminum cap and pressure-molded with a press of 15 to 20 tons to prepare a sample for analysis. The press molding method takes a shorter time to obtain a sample for analysis than the glass bead method, but requires two analyzes on the same sample in order to eliminate measurement errors due to variations in crushed particles. . Therefore, as a result, 2
It takes about 0 minutes.

In order to shorten the analysis time, the present inventors have included in the steelmaking slag based on the amount of oxygen discharged to the outside of the system due to the reaction between the sample collected from the steelmaking slag and the carbon source. A method for quantifying easily reducible metal oxides such as Cr, Fe and Mn was developed and filed as Japanese Patent Application No. 4-319364 on November 4, 1992. By this method, it becomes possible to quantify oxygen in the slag in a short time within 10 minutes. On the other hand, as a method of measuring the amount of slag, there is known a method of measuring the weight of the ladle for each tapped steel and slag and subtracting it from the total weight of the ladle including the amount of molten steel and slag, but it is still known. Is difficult to measure with high accuracy.

[0007]

Therefore, the present inventors have calculated the slag weight W _S from the total weight W _T , the molten metal weight W _M and the weight W _{V of the} vacuum degassing vessel, ladle, etc., which can be easily weighed. Developed a method to do. According to this method, the slag weight W _S is calculated as W _S = W _T − (W _M + W _V ). In this method, it is required to detect the molten metal surface and grasp the molten metal weight W _M. As a method of detecting the level of the molten metal surface under the slag, the difference in electrical resistance between the slag and the metal is used, the electrode is inserted through the slag, and the molten metal surface is detected by the change in electrical resistance. Can be considered. However, it is difficult to measure by this method because the solidified state of slag varies. On the other hand,
The use of an eddy current sensor can be considered as a means for measuring the molten metal surface in a non-contact manner.

Eddy current sensors are used to detect the level of molten metal in a continuous casting mold. When a high frequency current is supplied to the high frequency coil of the eddy current sensor, an eddy current is generated on the molten metal surface which is a conductor. An induced magnetic field is generated near the molten metal surface by the generated eddy current, and the induced magnetic field changes the impedance of the coil. The change in the impedance of the coil is converted into a voltage or a frequency and taken out as information indicating the molten metal level. The slag floating on the molten metal surface in the continuous casting mold is derived from an insulating flux based on oxides of alkali metals and alkaline earth metals. In this case, the eddy current generated by the high frequency current supplied to the high frequency coil of the eddy current sensor is specified on the molten metal surface, and the error factor due to the slag taken into the eddy current sensor can be ignored.

However, a large amount of metal particles are suspended in a droplet form in the slag floating on the surface of the molten metal contained in a refining vessel such as a converter, a ladle, and a vacuum degassing vessel. ing.
For example, in the slag layer inside the converter, metal droplets of about 0.1 to 10 mm may be suspended by about 50% by volume. If the eddy current sensor is used to detect the molten metal surface through the slag layer in which metal droplets, which are conductive substances, are suspended, the influence of the eddy current generated in the slag layer will be large and the reliability of the measurement results will be reduced. . The present invention has been devised to solve such a problem, and by specifying the frequency of the alternating current supplied to the eddy current sensor, a large amount of conductive metal droplets are suspended. The purpose is to accurately detect the molten metal level even through the slag layer.

[0010]

[Means for Solving the Problems] In order to achieve the object, the method for measuring the level of a molten metal of the present invention has a surface of a metal bath in which a slag layer in which 50% by volume or less of metal particles are suspended is suspended. At the time of measuring, the eddy current sensor is arranged above the metal bath, an alternating current having a frequency of 0.5 to 500 KHz is supplied to the eddy current sensor, and the eddy current is applied to the molten metal bath surface by the alternating current. It is characterized in that the level of the molten metal is calculated based on the induced magnetic field generated by the eddy current.

The measurement of the molten metal surface by the eddy current sensor is shown in FIG.
Will be described with reference to. As the eddy current sensor 10, the one in which the receiving coils 12 and 13 are arranged above and below the oscillation coil 11 is used in order to increase the measurement accuracy. When a high-frequency current is supplied to the oscillation coil 11 arranged near the molten metal surface 14, an eddy current 15 flows in the molten metal surface 14. Eddy current 1
5, an induction magnetic field 16 is generated, and magnetic field lines 17 are detected by the receiving coils 12 and 13. The strength of the eddy current 15 and, consequently, the strength of the magnetic field lines 17 detected by the receiving coils 12 and 13 vary depending on the distance between the eddy current sensor 10 and the molten metal surface 14. Therefore, the position of the molten metal surface 14 can be detected based on the magnetic field line strength detected by the receiving coils 12 and 13. It should be noted that either one of the receiving coils 12 and 13 can be omitted. Also, the receiving coils 12 and 13
It is also possible to provide the two in succession.

When metal droplets are suspended in the slag layer 18 floating on the molten metal surface 14, an eddy current is also generated in the slag layer 18. As a result, an induced magnetic field is generated due to the eddy current of the slag layer 18, and the error factor is taken into the receiving coils 12 and 13. The eddy current generated in the slag layer 18 can be ignored in the slag layer in which the amount of suspended metal droplets is small. However, a large eddy current is generated in the slag layer 18 in which the amount of suspension may reach up to about 50% by volume like the slag in the converter because it is electrically conductive. An induced magnetic field is generated by this eddy current, and an error factor with respect to the measured value is taken into the receiving coils 12 and 13.

The effect of the frequency of the alternating current supplied to the oscillating coil 11 on improving the accuracy of the measured value is not clear.
However, generally, the smaller the frequency of the supplied current, the thicker the skin depth of the magnetic field generated on the molten metal surface (distance in which the magnetic field decreases in the depth direction of the molten metal surface), and the reliability of the measured value decreases.
On the contrary, if the frequency is too large, eddy current is generated on the slag surface near the sensor side, and the sensor detects the slag surface. From this, 0.5-500
It was experimentally clarified that when the measurement is performed while supplying a high frequency alternating current of kHz to the oscillation coil 11, the molten metal level can be detected with high accuracy without being affected by the conductive slag layer 18. The measurement of the molten metal level using the alternating current of the specified frequency shows that the dispersion amount of the metal droplets in the slag layer 18 floating on the molten metal surface 14 is 50% by volume.
Is valid until. However, there are 50
In the slag layer 18 that exceeds the volume%, a large eddy current is generated in the slag layer 18 as the conductivity increases, and the influence of the induced magnetic field generated by the eddy current increases. Therefore, the reliability of the measured value is reduced.

A slag layer 1 suspended on the molten metal surface 14
In No. 8, the metal droplets 19 are not uniformly dispersed as schematically shown in FIG. Therefore, at the metal droplets 19 having a relatively large diameter or at locations where the metal droplets 19 are densely dispersed, an error factor due to the metal droplets 19 is taken in and the distance d ₁ measured by the eddy current sensor 10 is taken. May be shorter than the actual distance d ₀ to the molten metal surface 14. The measurement result of the distance d ₁ from the information indicating the molten metal level is excluded by scanning the molten metal surface with the eddy current sensor 10. For example, as shown in FIG. 4, an arm 22 is attached to a revolving shaft 21 provided around the container 10 so that the eddy current sensor 10 is located above the inside of the container 20, and the eddy current sensor 10 is attached to the tip of the arm 22. To provide. Then, the eddy current sensor 10 is swung in the arrow direction above the slag layer 18. The eddy current sensor 10 swings, and the molten metal surface 1
4 levels are detected continuously. When picking up the longest value from the obtained detection result, the short distance d ₁ caused by the metal droplet 19 shown in FIG. 3 is excluded.

[0015]

Example 1 (Experiment using pseudo slag) Cu particles having an average particle size of 1.0 mm were dispersed in slag powder at a ratio of 20% by volume to form a rectangular parallelepiped shape having a length of 100 mm, a width of 100 mm and a height of 25 mm. Molded into. A slag molded body was arranged on a stainless steel plate SUS304 which was a non-magnetic material, and an eddy current sensor was provided above the slag molded body so as to be able to move up and down. Then, an alternating current of 50 kHz was supplied to the eddy current sensor, and the distance d from the surface of the stainless steel plate to the eddy current sensor was measured via the slag forming body. Also,
The eddy current sensor was arranged above the stainless steel plate without interposing the slag forming body, and the distance d ₂ to the surface of the stainless steel plate was measured by the eddy current sensor.

The distances d and d ₂ obtained by the eddy current sensor
Was compared with the actually measured distance D from the surface of the stainless steel plate to the eddy current sensor. As a result, the distances d and d ₂ were substantially the same value regardless of the presence or absence of the slag molded body, and the distances d and d ₂ matched the distance D within a range of ± 1%.
From this, it is understood that the magnetic field lines of the induced magnetic field due to the Cu particles in the slag are not taken in by the eddy current sensor as an error factor that reduces the reliability of the measured value.

Example 2: (Effect caused by the amount of dispersed metal particles) The amount of Cu particles dispersed in the slag compact was changed and the measured value D was 60.
The distances d and d ₂ to the surface of the stainless steel plate were measured under the same conditions as in Example 1 except that the value was maintained at a constant value of mm. The measurement result d is compared with the actual measurement value D, and the measurement error δD [=
The relationship between (d−d ₂ ) / D] and the amount of Cu particles dispersed in the slag compact was investigated. As is clear from Table 1 showing the investigation results, in the measurement through the slag molded body in which Cu particles of 50% by volume or less are dispersed, the distances d and d are the same as in Example 1.
₂ was in good agreement with the measured distance D. However, Cu
If the amount of dispersed particles exceeds 50% by volume, the distance d ₂
The distance d tended to be smaller than that of

[Table 1]

Example 3 (Influence by Frequency) Example 3 was repeated except that the measured distance D from the surface of the stainless steel plate to the eddy current sensor was maintained at a constant value of 60 mm and the frequency of the alternating current supplied to the eddy current sensor was changed. The distances d and d ₂ were measured under the same conditions as in Example 1 and compared with the measured distance D. As is clear from Table 2 showing the comparison result, it is understood that the measurement can be performed with extremely high accuracy of the error δD of ± 2% in the frequency range of 0.5 to 500 kHz. On the other hand, in the measurement in which the alternating current having the frequency of 0.1 kHz was supplied, both the distances d and d ₂ tended to be long. It is speculated that this is because the eddy current generated on the surface of the stainless steel plate is weak. On the other hand, in the measurement in which the alternating current having the frequency of 1000 kHz was supplied, the distance d tended to be shortened. It is presumed that this is because the error factor due to the eddy current generated in the slag molded body was taken into the receiving coil.

[Table 2]

Example 4: (Measurement Test) From the results of Examples 1 to 3, the metal particle dispersion amount was 50% by volume.
Even if the following slag layer is interposed, the frequency is 0.5 to 500.
It is understood that the surface position of the stainless steel, which is a conductor, can be measured with extremely high accuracy when measuring the length while supplying an alternating current of kHz to the eddy current sensor. Based on this experimental result, a test using a melting furnace was performed. As a melting furnace,
A 500 kg laboratory furnace with a refractory lining was used. Into this experimental furnace, 500 kg of SUS304 molten stainless steel was charged. 500mm from the bottom of the furnace
Located at the height of. The internal cross-sectional area of the experimental furnace was 0.2 m ² at the position where the molten metal surface was located. In this state, that is, in a state where the slag is not on the surface of the molten metal, the distance d ₂ is obtained by the eddy current sensor arranged above the molten metal surface
The optical system measured the distance D from the molten metal surface to the eddy current sensor.

The molten stainless steel was maintained at a temperature of 1600 ° C., and 50 kg of silica-alumina slag was charged on it. As the slag, SiO ₂ : 40% by weight, Al ₂
The basic composition was O ₃ : 10 wt% and CaO: 40 wt%, and the ferrochrome suspended as metal droplets was changed in various proportions. The slag formed a slag layer that was evenly distributed over the entire molten metal surface 10 minutes after the charging. The calculated thickness of the slag at this time is 100 mm.

An eddy current sensor was arranged at a height of a measured distance D from the molten metal surface, an alternating current having a frequency of 50 kHz was supplied, and the distance d was obtained by the eddy current sensor. When the relationship between the length measurement result and the ferrochrome content in the slag was investigated,
The measurement error δD (%) represented by (d−d ₂ ) / D is shown in FIG.
As shown in (3), the tendency of the ferrochrome particles to increase with the increase in the suspension amount was shown. However, in the measurement of the level of the molten metal surface through the slag layer in which the suspended amount of ferrochrome particles is 50% by volume or less, the measurement error δd is less than 2%, which is extremely high, regardless of the measured distance D from the molten metal surface to the eddy current sensor. It was possible to measure the molten metal level with accuracy. This result is consistent with the measurement result of Table 1 using the pseudo slag.

Example 5 (Calculation of Slag Amount) Stainless steel SUS304 was smelted with an experimental vacuum degassing apparatus having a capacity of 500 kg. After refining, the container containing molten steel and slag was weighed, and the total weight W _T was 12
It was 50 kg. It is known that the weight W _{V of the} container itself is 700 kg by weighing before refining. After the level of the molten metal and the slag layer had subsided, the level of the molten metal was measured with an eddy current sensor arranged above the slag layer. The level of the molten metal was found to be 500 mm above the bottom surface of the container. The weight W _M of the molten steel calculated from the position of the molten metal surface was 500 kg. Therefore, the weight W _s of the slag floating on the surface of the molten steel is 5 from the relationship of W _s = W _T − (W _M + W _V ).
It was calculated to be 0 kg.

Next, molten steel was tapped from a nozzle provided on the bottom wall of the container. At this time, a nozzle having a built-in electrode for detecting the conductivity of the fluid passing through the nozzle was used, and a method was adopted in which the nozzle was immediately closed when the conductivity was decreased by the electrode. As a result, the molten steel and the slag could be discharged from the container in a completely separated state. When the discharged molten steel and slag were weighed, they were 498 kg and 49 kg, respectively. These measured values were extremely close to the calculated values W _M and W _S. From this example, it can be seen that when using the eddy current sensor according to the present invention, accurate information about the molten metal surface and the slag layer after the converter refining, vacuum degassing, etc. is obtained. The information on the obtained amount of slag can be obtained by, for example, determining the amount of reducing agent addition required when reducing easily reducible metals such as Fe, Mn, and Cr contained in slag with Si etc. used. Further, the molten metal weight is calculated from the measured molten metal surface level, and it becomes possible to predict the molten metal discharge time.

[0024]

As described above, according to the present invention, the molten metal level can be accurately detected even through the slag layer in which the metal droplets are suspended. Therefore, the weight of the molten metal, the weight of the slag, and the like can be accurately grasped, and the operation management of the molten metal, the molten metal, and the like can be accurately performed. Further, since the slag layer floating on the molten metal surface is also calculated from the detected molten metal level, the reduction added when recovering valuable metals such as Fe, Mn and Ni contained in the slag to the molten metal. It is possible to accurately adjust the amount of the agent added.

[Brief description of drawings]

FIG. 1 Flow chart showing a method for quantifying oxygen in a metal oxide

FIG. 2 is an explanatory diagram for measuring the molten metal surface with an eddy current sensor.

[Figure 3] Slag floating on the surface of the bath

[Fig. 4] Eddy current sensor for scanning the molten metal surface

FIG. 5: Effect of metal droplets in slag on measurement error

[Explanation of symbols]

10: Eddy current sensor 11: Oscillation coil 12,
13: receiving coil 14: molten metal surface 15: eddy current
16: Induction magnetic field 17: Magnetic field lines 18: Slag layer
19: Metal droplet 20: Container 21: Swivel axis
22: Arm

Claims (1)

[Claims] 1. When measuring the level of a metal bath in which a slag layer in which 50% by volume or less of metal particles are suspended is placed, an eddy current sensor is arranged above the metal bath, and a frequency of 0.
An alternating current of 5 to 500 KHz is supplied to the eddy current sensor, an eddy current is generated on the molten metal surface of the metal bath by the alternating current, and the molten metal level of the molten metal is adjusted based on an induction magnetic field generated by the eddy current. A method for measuring the level of a molten metal surface under a slag, which comprises calculating.