FIELD OF THE INVENTION
This invention generally relates to a methodology for assessment of cleanliness of continuously cast steel slabs. Particularly the invention relates to a methodology for categorizing the quality of continuously cast steel slabs based on cleanliness measurement using contact ultrasonic technique. More particularly, the invention relates to a method for classification of continuously cast steel slab with respect to quality using ultrasonic technique.
BACKGROUND OF THE INVENTION
Continuous Casting is a process whereby molten steel is solidified into a slab for subsequent rolling in the rolling mills in order to get the finished products. Fig. 1 shows a typical slab caster layout. Liquid steel from the basic oxygen furnace is tapped into a ladle and taken to a continuous casting machine. The ladle is brought onto a turret that rotates the ladle into the casting position above the tundish. Tundish is the temporary buffer of liquid steel, which feeds water, cooled oscillating mould. Solidification begins in the mould, and continues as a strand which is straightened, torch-cut and then discharged for intermediate storage or direct hot rolling. The final products from the slabs are sheets. Sheets rolled from prime grade (L1 quality) slabs should be the cleanest in terms of inclusion content, voids and other defects, and are used for automotive manufacture.
Producing defect free steel sheets is very important as the automotive manufacturing demand for very high surface quality. Typical discontinuities are slivers and cracks, which are all the result of non-metallic inclusions and voids presents in the slabs. Meeting customer requirement is the key for success in any business and hence supplying clean steel is very important. The challenge

has two faces; one is producing clean steel and the other is measuring the quality (cleanliness) of the produced clean steel. High quality of steel is achieved by optimized process parameters such as mold level fluctuation, tundish weight, stopper rod position, casting speed etc. Any deviation in the process parameters can potentially turn into an origin of discontinuities in slabs. Process deviations are used as a yardstick while categorizing the slabs into different qualities. The norms for categorizing the slabs are generally established based on statistical study for a particular mill. However there are outliers in statistics, which leads to misclassification of many slabs. In this context, this invention enables slab classification based on ultrasonic testing technique to avoid such misclassifications. The invention is related to a methodology which involves ultrasonic testing of slabs and interpretation of ultrasonic results for classification of slabs.
Slab cleanliness measurement is quite difficult because all the existing methodology involves assessing the quality of the slabs by destructive methods on samples basis and/or analysing the surface condition using online image processing.
IN 260658 teaches a method of assessment of cleanliness level of continuously cast steel by an automatic ultrasonic immersion C-scan image analysis. The aim of this work was to assess the cleanliness level of continuously cast steel by an automatic ultrasonic immersion C-scan image analysis on sample basis. This method involves preparing cylindrical samples from ladle furnace as well as tundish, cutting and grounding the samples to mirror finish; employing ultrasonic scanning in immersion tank with water and the scanner using 5 MHz and 10 MHz focused beam probe. This patent claims that the quality of the sample of continuously cast steel is assessed by correlating total oxygen with ultrasonic results.

US5884685 A describes a quality prediction and quality control of continuous cast steel using mathematical model. The objective of this work was to provide a method of continuously calculating a non metallic inclusion distribution at an outlet of ladle and tundish using mathematical model supplied with operational data.
US 8499821 B2 discloses a method for detecting and classifying surface defects on continuously cast slabs using topographical information which is about the appearance of continuously cast surface. It describes that surface topography of the slabs is determined by either optical or laser based methods. The evaluation of the detected changes in topography can be interpreted by neural networks.
US4519041 A describes a method for real time automated inspection of moving hot slabs. The aim of this work is to device a camera which scans slab continuously in transverse direction in order to sense light intensities and generates voltage values. The voltage values are converted to corresponding digital values to form a digital image of the surface using edge enhancement filter.
All the prior art discussed hereinabove, are intended for assessing the quality of slabs either using mathematical model, optical or laser based methods and through image processing except IN 260658. This patent assesses the cleanliness level on sample basis which is not a representation of entire slab. The invention is focused on assessing the quality of slabs based on the samples collected from ladle and tundish. It is not focused on the entire final product so called continuously cast steel slabs.

A method for classification of continuously cast steel slab with respect to quality using untrasonic technique.
OBJECT OF THE INVENTION
It is therefore the object of the invention to propose a method for classification of continuously cast steel slab with respect to quality using untrasonic technique.
SUMMARY OF THE INVENTION
Accordingly, there is provided a method for classification of continuously cast steel slab with respect to quality using untrasonic technique. Thus, the present invention is directed to measuring the cleanliness of the cast steel slabs. Because the quality of the slabs is categorized based on statistical study which involves the measurement of different processing parameters like mold level fluctuation, tundish weight, stopper rod position, casting speed etc, during casting and comparison of these parameters with generally formed sliver defect, in coils after cold rolling. This invention is developed to assess the quality of the slabs after casting using ultrasonic testing. As mentioned in the prior art section, all the previous inventions employ various methods to assess the quality of the slabs like optical or laser, mathematical model or through image processing.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
Fig. 1 – Slab Caster Layout
Fig. 2 – Schematic drawing of Calibration Block
Fig. 3 – Calibration Block with Flat Bottom Holes
Fig. 4 – Ultrasonic echo amplitude for ø 2 mm FBH at 6mm depth

Fig. 5 – Calibration Curve using Calibration Block.
Fig. 6 – Schematic of the Grid made for Ultrasonic Testing.
Fig. 7 – Comparison of Different IF grade slabs.
Fig. 8 – Distribution of Normalized Ultrasonic Results L2 Slabs vs MLF.
DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT OF THE INVENTION
Importance of Slab Quality Assessment
The interstitial-free (IF) steel is widely used in automotive applications especially for exposed panels which require the control of non-metallic inclusions and good surface quality. Because inclusions deteriorate the mechanical properties of steel under tension, bending, hole expanding, press forming, and other types of working which will eventually cause surface defects. High quality products will be produced if the slabs are defect free. The slabs are categorically divided into L1, L2, L3 and L4 grades (quality in descending order) depending upon the different casting parameters like mold level fluctuation, stopper position, change in casting speed and tundish weight etc., The significance of these parameters are as follows;
Mould Level Fluctuation (MLF)
Mould level control is a function of the process control of continuous casting to maintain the constant level of molten steel surface in a mould (hereafter it is referred as the mould level). This has a significant influence over the quality and yield of the final product. There is a close correlation between the fluctuation of the mould steel level and the occurrence of surface defects in final products. Mould level fluctuation can cause entrapment of casting powder and other impurities floating on the surface of molten steel and they

appear in the form of surface defects of steel sheet products during rolling. To prevent this, mould level control is designed so as to minimize the fluctuation of the mould steel level.
Stopper Rod Position
The sliver defect formation and quality of the final steel product are closely related to the transient liquid steel flow in the mould. Transient flow in the submerged entry nozzle (SEN) and mould is strongly affected by the changing SEN flow rate, which is eventually controlled by the stopper rod position. Abrupt movements of the stopper rod are known to cause sliver defects.
Casting Speed
Casting speed is an important parameter for producing quality of slabs by continuous casting. If the casting speed is high, it may cause inadequate thickness of the solidified shell at the mould. It leads to breakout in the line. Breakout is one of the operation troubles that most adversely affect the production of continuous casters. In addition high casting speed also causes entrapment of non-metallic inclusions during casting. On the other hand, an excessively slow casting speed causes a loss of production.
Tundish Weight
Tundish weight indicates the amount liquid steel available in tundish for getting fed to mould for casting. Minimal tundish weight should be maintained to avoid vortex formation and subsequent slag entrapment during casting. Low tundish weight happens when there is a delay in ladle change.

The norms for classifying the slabs are established based on statistical study. This study involves measurement of different processing conditions during casting, as mentioned above, and comparison of these parameters with slivers in coils after cold rolling. But the statistical study based methodology allows misclassification of many slabs, as every statistics has some outliers. So it is very much needed to develop a methodology, which measures the quality of all slabs and avoids misclassifications.
Methodology for the slab quality
In order to utilize ultrasonic testing as a slab quality assessment tool, testing procedure needs to be optimized. For this purpose a calibration block with known discontinuities which simulate actual ultrasonic responses of slabs is required.
Calibration Block
A calibration block was made from a sample of IF grade (L1) size of SQ 28 x 210 mm long. The block has six Flat Bottom Hole (FBH) of Ø2.0 mm with the depths of 1 mm, 2 mm, 4 mm, 6 mm, 8 mm and 10 mm, respectively as shown in Fig. 2. Actual calibration block photograph is depicted in Fig.3
Ultrasonic Inspection System
The pulse-echo technique is a type of ultrasonic method where a short pulse or burst of ultrasonic energy is transmitted from a probe into a material. If this energy is met by a discontinuity such as an inclusion or pin holes, part of the energy is reflected back to a search unit as an echo. It is then converted to electrical energy and displayed as a pulse on an oscilloscope screen. Another burst of energy is then sent into the material and the whole process repeated. The time interval between bursts is long enough to allow the echo

to return. The repetition rate of this burst can be adjusted from 60 to 2000 Hz.
Ultrasonic probes which transmit high frequency ultrasound can resolve smaller discontinuities, but has lesser penetration capability. Meanwhile low frequency probes are less attenuated by the specimen, but are also less sensitive to discontinuities. In order to evaluate the quality of the slabs which include both surface and subsurface (10 mm depth), 6MHz Twin Crystal Transmitter-Receiver (TR) probe was selected optimally.
Calibration block was tested using a known ultrasonic equipment Fig. 4 shows the ultrasonic echo amplitude of ø2 mm FBH at 6 mm depth. During testing the gain of the instrument was set as 63 dB optimally, which should be maintained during testing of actual slabs. Calibration curve for 2 mm hole at different depth is obtained from the testing and plotted as shown in Fig. 5.
Normalisation of Calibration Curve
Size and depth of the discontinuity influences the amplitude of ultrasonic echo. It is a known fact that the size of the discontinuity and ultrasonic amplitude has linear relationship; ultrasonic amplitude can be a tool for assessing the discontinuity size. However, the effect of depth on the amplitude should also be incorporated. Here depth is affecting the amplitude by two means; one is focal effect of the ultrasonic probe (till 6 mm depth) and the other is attenuation of ultrasonic wave. But the ultrasonic study of slab is based on the severity of discontinuities, which is the size of the defect alone. Hence the calibration curve should be normalised to eliminate the effect of depth. Normalisation was done with respect to 6 mm depth and the normalisation factors are calculated as given in Table 1.


With the above calibration, the following procedure is followed.
The testing was done at 100 points for each slab, which is sufficiently large to assess the quality of the slab as shown in Fig. 6. As mentioned in the calibration section, ultrasonic probe of 6 MHz was used in testing. Range of ultrasonic testing was set 40 mm and the gain was set 63 dB. The test results were recorded as ultrasonic amplitude against its depth value. The ultrasonic echo amplitude obtained at 5 mm and 6 mm depths were shown in Fig. 8 and 9, and the ultrasonic amplitude of the results is 17 and 15% FSH (Full Screen Height), which is comparable with 2 mm FBH in the calibration block. Hence it proves the closeness of the calibration block with actual slab conditions.

Results
Normalisation of Ultrasonic Test Data
The obtained data has two variables: depth and amplitude. For eliminating the depth variable, the amplitude should be multiplied by the respective normalization factor given in Table 1. Normalised ultrasonic amplitude can be read for the severity of the discontinuities. Table 2 shows few ultrasonic results of the slab ultrasonic testing result and its normalised amplitude for the understanding.


As the normalized ultrasonic data for each slab is found to be distributed to maximum of 38% FSH, the ultrasonic data were analysed as plotted. The test results are shown in Fig. 7, which shows the distribution of ultrasonic readings for different categories of IF slabs. The box in the each reading shows the position of the first and third quartile of the sample population, i.e. the position of the middle 50 % population. So from the box of L1 slabs, it is inferred that the normalized ultrasonic amplitude of 12.852 % FSH can be taken as the acceptance limit for slab quality to consider as L1. All the slabs can be tested as per above methodology and the test readings can be compared against this limit for categorizing it as L1 or L2. This way all the slabs can be categorized as per its cleanliness rather than based on the process parameters. Ultrasonic testing of all slabs can be done using automated ultrasonic inspection.

WE CLAIM :
1. A method for assessing the quality of steel slab, the method comprising steps of:
determining various no. of defect size in terms of ultrasonic echo amplitude present in various slab samples along with their corresponding depth from the scanning surface, the slab being made of a process;
calculating the normalization factor for eliminating depth effect on amplitude of the ultrasonic echo, the normalization factor (Nx)
NX = AFBHxI
Af
where AFBHX is the depth and Af is the ultrasonic echo amplitude;
multiplying each amplitude with the normalization factor, to generate normalized amplitude, the normalized amplitude represents the defect severity irrespective of its location (depth);
distributing the normalized amplitude for the various slabs, thereby generating a master graph; and
determining defect size and multiplying with normalization factor of a fresh steel made by the said process and further distributing over the master graph and to assess the quality.