Method of monitoring effervescence of a steel

Abstract

The sound emitted by steel being teemed into an ingot mould and by the steel solidifying in the ingot mould is detected. The acoustic energy of emitted sound of at least one frequency in a frequency range from 0 to 100 kHz is measured and recorded as a function of time. For instance the acoustic energy at one frequency may be recorded during teeming and acoustic energy at two other frequencies may be recorded during subsequent effervescence. The energy measurement record is compared with results obtained by statistical analysis if similar records resulting from monitoring of a large number of previous ingots. This comparison reveals one or more characteristics of the sound emitted which give an indication of the quality of the effervescence of the steel. If the quality indicated is inferior, it is possible to correct the composition of the steel in the ingot mould during teeming or soon after.

Description

The present invention relates to a method of monitoring the effervescence (rimming) of a steel during teeming and solidification in an ingot mould.

Deoxidation by the carbon of unkilled (rimming or balanced) steel results in the well known phenomenon of rimming (effervescence) during solidification of the steel in ingot moulds. This phenomenon occurs as boiling of the molten upper surface of the steel ingot, due to violent evolution of gas substantially consisting of CO. This boil is also accompanied by projection of sparks (incandescent metal particles).

Many factors influence the phenomenon of rimming, such as the temperature of the steel and the ingot mould, the solidification rate of the metal, its composition, the proportion of any deoxidizing elements (e.g., manganese and silicon) and the shape and dimensions of the ingot mould.

For a long time assessment of rimming was left to the experience of the observer, who used to classify the ingots in various categories on the basis of a general evaluation of the effervescence, and who of course could rely only upon his own substantially subjective impressions.

It should be noted that such a method has remarkable drawbacks, particularly in view of the risk of errors in assessment, variations in assessment depending on the tiredness of the observer, and the non-reproducibility of the results because of the substantially subjective nature of the observation means, i.e., the eye of the observer, and the observation method, i.e., data of personal experience stored in his memory.

In order to evaluate the gas evolution rate more objectively, a measuring method has been suggested comprising recording all or a part of the sound spectrum (sonic and ultrasonic) emitted by the steel in the ingot mould.

It has actually been noted that a relation exists between the evolution of the phenomenon observed and the evolution of its sound spectrum, which allows some indications to be obtained with regard to such a phenomenon by means of measurements taken on its spectrum.

Such a method, although it has led to satisfactory results, has the drawback of obtaining an assessment of the effervescence only at the end of teeming, which provides the possibility of only tardy corrections.

The present invention concerns a method which in particular allows this drawback to be avoided.

The invention is based on the remarkable observation that the teeming noise of rimming steels already contains indications of effervescence, which makes it possible to get information from the very first beginning of the teeming operation and to make any necessary corrections very rapidly and efficiently.

Another observation is that the solidification process of steel produces two noises of different frequencies: one noise has frequencies lower than 10 kHz due to vortices in the molten metal, and another noise has frequencies higher than 10 kHz due to bursting of bubbles of gas evolving from the molten metal and to projection of sparks. These two phenomena occur during solidification of steel and as a result of its effervescence. Experience has suggested that it is important to monitor both of them, preferably at the same time, in order to better assess the quality of the process taking place.

In the method according to the present invention the acoustic energy (sonic and ultrasonic) from the steel while being teemed into ingot mounds and while solidifying into an ingot is recorded as a function of time and within a frequency range from 0 to 100 kHz (preferably from 0 to 40 kHz) and the effervescence rate and possibly the materials to be added, for correcting this effervescence as a function of specific characteristics of the acoustic energy, are determined by comparison with results obtained from a statistic study based on a large number of ingots. Acoustic energy can be picked up by an acoustic transducer or an accelerometer.

If the characteristics indicate an inferior quality of effervescence, this can be corrected during teeming or soon after, by adding oxidising agents or deoxidants.

Suitable characteristics of the emitted sound can comprise the form of the spectrum and/or development of the frequency components of the spectrum and/or the relative or absolute level of these components and/or the amplitude of their fluctuations.

For example, when picking up the sonic energy during teeming of steel, the quality of the effervescence (and possible the materials to be added for correcting this effervescence) is determined as a function of the average level and/or the rate of increase of the time curve of the evolution of the sonic energy filtered at at least one given frequency.

It has been observed in fact that at a high average level of the sonic energy a good start of the effervescence occurs, while at a not very high average level effervescence starts badly.

In the case in which the quality of the effervescence is determined as a function of the average level of the sonic energy, it has been found to be particularly advantageous to observe the time evolution of the sonic energy filtered to a frequency of 8 kHz approximately.

When the quality of the effervescence is determined as a function of the rate of increase of the sonic energy, the time evolution of the sonic energy filtered in respect of a frequency of 18 kHz advantageously observed.

The picked up sonic energy not only depends upon the metallurgic process (appearance of effervescence during casting) but also upon external factors such as the form and dimensions of the ingot mould, and the position of the picking up device. In order to eliminate the influence of these other factors, the picked up sonic energy is filtered at a first frequency which is affected by the effervescence and the external factors, such as at 8 kHz, and also at a second frequency mainly affected by the external factors, at 1 kHz example, and the ratio between the two sonic energies thus filtered is determined to obtain a purely representative index of the quality of the effervescence in the steel being tested.

In the case in which the acoustic energy during teeming is picked up, the quality of rimming (and possibly the materials to be added for correcting (effervescence) can be determined by filtering the acoustic energy to a first frequency between 1 kHz and f.sub.1 (where f.sub.1 is betwen 2 and 8 kHz) and to a second frequency between f.sub.1 and 10 kHz, measuring the acoustic intensity at these two frequencies, and comparing the data at these two frequencies with predetermined values.

The indications thus obtained are employed to forsee the behaviour during rimming and to decide in a very rapid manner on the possible corrections to be made (e.g., addition of oxidizing powder if rimming is to be increased, or addition of deoxidants if rimming is to be decreased).

In the case in which the sonic energy from the effervescence during rimming of the steel in the ingot mould is picked up, the effervescence quality (and possibly the materials to be added for correcting this effervescence) is determined by filtering the sonic energy to a first frequency between 1 kHz and f.sub.2 (f.sub.2 being between 5 kHz and 15 kHz) and to a second frequency between f.sub.2 and 40 kHz, measuring the acoustic intensity at these two frequencies, and comparing the data at these two frequencies with predetermined values.

It has been found to be advantageous to record the time evolution of the sonic energy filtered to a first frequency of 5 kHz and a second frequency of 18 kHz.

Frequencies have to be chosen in accordance with the equipment used and the environment, in order to avoid the room noise and particular noise such as metallic impact and bridge crane noise.

In this case there is observed that a regular time evolution (that is an evolution at a relatively uniform level) shown by the curves of the sonic energy filtered to 5 kHz and to 18 kHz, corresponds to satisfactory effervescence while a deficient effervescence occurs as a result of an irregular evolution, i.e., an evolution at a variable level.

It should also be noted that the long term evolutions of the sonic energy filtered to these two frequencies are different from one another and characteristic of the effervescence. Thus when the boil of the metal bath becomes irregular (wild) which gives an indication of a deficient effervescence, the level of the noise due to vortices in the metal bath (low frequency) changes by undergoing an increase, while the level of noise due to the bubble burst and the projection of sparks (high frequency) varies by undergoing a decrease.

On the other hand, when the boil of the metal bath is slow, which is another indication of a deficient effervescence, the level of noise due to vortices in the metal bath and to the bursting of the gas bubbles decreases.

EXAMPLES

The invention will be described further, by way of example only, with reference to the accompanying drawings, in which:

FIGS. 1 to 4 are graphs of acoustic energy against time, at various frequencies.

The graphs show the evolutions as a function of the time selectively filtered components (at 5 kHz, 8 kHz and 18 kHz) of the noise emitted by steel while being teemed (FIGS. 1 and 2) and during effervescence (FIGS. 3 and 4). The amplitudes of these components are shown in a logarithmic scale on the vertical axes (y). Time from the beginning of teeming is recorded linearly on the abscissa axis (t).

The observations described below relate to two rimming steel ingots obtained from two different heats. The first was an ingot of 16.5 tonnes, case in size 1,200 (analysis: carbon 0.064% and manganese 0.300%; temperature of the steel in ladle: 1,550.degree.C). The second ingot was an ingot of 18 tonnes, case in size 1,400 (analysis: carbon 0.075% and manganese 0.300%; temperature in the ladle: 1,542,20 C).

The solidification process of the first ingot is an example of good behaviour: quick commencement of rimming after teeming, no immediate rise or reboil, good and regular rimming until the ingot is hardened at its surface; and horizontal closing of the head.

In contrast, the second ingot is an example of bad behaviour: slow starting and steel reboiling in the first few minutes after teeming; slow and disorderly boiling; and bad closing.

FIG. 1 shows the evolution of the component at 8 kHz of the noise emitted, during teeming and for 40 seconds after teeming, by the first ingot (curve 1) and by the second ingot (curve 2).

The origin of the time axis corresponds to the beginning of teeming. The end of the teeming is indicated in the graph by a drop 3 in the two curves; the remaining part of the graph shows the noise emitted by the rimming metal bath during the first period immediately after teeming. Three short returns of the jet during the second ingot correspond to peaks 4 in the curve 2.

It should be noted that the level of curve 1 for the first ingot is, from the beginning, higher than the curve 2 for the second ingot; the average difference between the two curves is 15 decibels. This difference is due to a more substantial effervescence occurring during teeming of the first ingot than that appearing during teeming of the second ingot. This allows one to foresee from the very first minutes of teeming that rimming will commence rapidly and will be lively, in the case of the first ingot, and that it will start slowly and will be sluggish in the case of the second ingot.

The first ingot does not need any correction, while the second ingot has to be corrected by addition of an oxidising powder.

The same conclusions are reached by examination of FIG. 2, showing the evolution of the component at 18 kHz of the teeming noise on the same scale of coordinates as in FIG. 1. The shape of the two curves is different: the most important appearances of effervescence correspond to a steeper rise and a higher level of the curve 1 in connection with the first ingot.

FIGS. 3 and 4 show the evolution of the effervescence noise, selectively filtered at 5 kHz (curve 1) and at 18 kHz (curve 2), during a 20 minute period from teeming to covering of the ingot. The origin of the time axis corresponds to the beginning of teeming; the end of teeming is indicated at 3. FIG. 3 relates to the first ingot. In this graph the two curves have a regular shape. They show short-term fluctuations around a relatively constant average level; such curves indicates a good lively and regular effervescence. FIG. 4 relates to the second ingot. The two curves have an irregular average level, which indicates disorderly effervescence. After the fourth minute, the curve 1 rises in a remarkable way, while the curve 2 passes through a minimum. Such a curve indicates a phase of wild effervescence during which the eddies or vortices become violent and disorientated, while spark projection decreases. This behaviour is accompanied by rising and sloping and results in bad solidification.

Claims

I claim:

1. A method of monitoring the effervescence of steel teemed into an ingot mould, comprising the steps of:

a. detecting the sound emitted by the steel being teemed and by the steel solidifying in the ingot mould;

b. measuring the acoustic energy of emitted sound of at least one frequency in a frequency range from 0 to 100 kHz;

c. recording the energy measurement as a function of time; and

d. comparing the energy measurement record with results obtained by statistical analysis of similar records resulting from monitoring of a large number of previous ingots, the comparison revealing at least one characteristic of the sound emitted, the characteristic giving an indication of the quality of the effervescence of the steel.

2. A method as claimed in claim 1, in which the said emitted sound of at least one frequency is in a frequency range from 0 to 40 kHz.

3. A method as claimed in claim 1, in which the said at least one characteristic is selected from the group consisting of the form of the spectrum of the emitted sound, the evolution of the acoustic energy of emitted sound of at least one given frequency, the absolute or relative acoustic energy level of emitted sound of at least one given frequency, and the amplitude of fluctuations in the acoustic energy of emitted sound of at least one given frequency.

4. A method as claimed in claim 1, in which the measuring step comprises measuring the acoustic energy of emitted sound of at least one given frequency during teeming, the characteristic of the sound being the average level of the acoustic energy during teeming.

5. A method as claimed in claim 4, in which a higher average level represents a better quality of effervescence than a lower average level.

6. A method as claimed in claim 4, in which the said given frequency is 8 kHz approximately.

7. A method as claimed in claim 1, in which the measuring step comprises measuring the acoustic energy of emitted sound of at least one frequency, during teeming, the characteristic of the sound being the rapidity of the increase in the acoustic energy during teeming.

8. A method as claimed in claim 7, in which the said given frequency is 18 kHz approximately.

9. A method as claimed in claim 1, in which the measuring step comprises, during teeming, measuring the acoustic energy of sound of a first frequency between 1 kHz and a frequency f.sub.1, where f.sub.1 is between 2 and 8 kHz, measuring the acoustic energy of sound of a second frequency between the frequency f.sub.2 and 10 kHz, and comparing the two measurements with predetermined values.

10. A method as claimed in claim 9, in which the said first frequency is 1 kHz.

11. A method as claimed in claim 9, in which the said second frequency is 8 kHz.

12. A method as claimed in claim 1, in which the measuring step comprises, during effervescence, measuring the acoustic energy of emitted sound of a first frequency between 1 and a frequency f.sub.2, where f.sub.2 is between 5 and 15 kHz, measuring the acoustic energy of emitted sound of a second frequency between f.sub.2 and 40 kHz, and comparing the measurements with predetermined values.

13. A method as claimed in claim 12, in which the said first frequency is 5 KHz.

14. A method as claimed in claim 12, in which the said second frequency is 18 kHz.

15. A method as claimed in claim 12, in which a substantially constant level of the two energy measurements represents a satisfactory quality of effervescence, an unsatisfactory quality being represented by a variable level.

16. A method as claimed in claim 12, further comprising the step of comparing the long term evolution of the energy measurement at the first frequency with that at the second frequency.