JPS5834774B2 - Method for detecting surface flaws on hot steel materials - Google Patents

Description

[Detailed description of the invention] The present invention relates to a method for detecting surface flaws on hot steel materials.

Conventionally, various methods have been proposed for online detection of surface flaws in hot steel materials such as hot steel slabs, billets, beams, and blanks immediately after blooming or continuous casting.

In addition, when carrying out this type of surface flaw detection method, impurities such as scale and water droplets on the surface of the hot steel material are removed to obtain a clean surface condition, and the surface is made to reveal flaws and make them easier to detect. Various processing methods have also been proposed.

For example, by spraying a fluid at a temperature lower than the surface temperature on the surface of a hot steel material to temporarily cool it, defects can be made apparent during the recuperation process, or by passing a high-frequency current through the hot steel material, the surface can be temporarily cooled. There is a method of heating the material to an even higher temperature to make the flaws appear during the subsequent cooling process.

All of these methods can effectively make the eaves visible and enhance the contrast with the background, so they are advantageous for improving defect detection quality.

However, on the other hand, since the surface of the hot steel material is more chemically activated, applying heat disturbance from the outside will promote surface oxidation.

On the other hand, in order to make deep scratches on the surface of heated steel materials whose surface temperature is over 1000℃ and which has been cleaned by applying solvent to the entire surface using a hot scarfer, etc., without giving the above-mentioned thermal disturbance, it is necessary to remove the scale immediately after applying the solvent. begins to occur and grows rapidly within 10 seconds. Therefore, unless deep damage occurs within 30 seconds after the solvent is applied, it is extremely difficult to distinguish between scale and defects such as bald spots.

As mentioned above, when heat disturbance is applied from the outside, surface oxidation is promoted and this tendency becomes even more pronounced, and the scale that is generated also becomes noticeably thicker and larger, so it is important to detect defects as early as possible after heat disturbance is applied. If this is not done, the flaws will become trapped in the scale and become undetectable.

In fact, when surface flaws are detected online by applying thermal disturbances, the time during which flaws can be clearly observed is within 15 seconds after the surface of the hot steel material is temporarily cooled or heated.

However, even if one attempts to obtain surface information of hot steel materials while satisfying the above-mentioned time conditions, it is extremely difficult to scan the surface in a state completely free of surface impurities such as scale and water droplets. The surface information has an extremely poor S/N ratio and is contaminated with signals from surface impurities.

These noise signals include those whose signal level is lower than the signal level of the healthy portion, that is, the background portion, and those whose signal level is higher than the signal level of the healthy portion, that is, the background portion, and generally there are overwhelmingly many low-level signals.

This is mainly due to secondary scales that rapidly grow to a certain size on the surface of the hot steel material.

Furthermore, among the flaw signals included in the surface information, there are many small flaw signals that can be removed in downstream processes without any flaw care.

For example, if there is a small projecting flaw such as a scale-like flake that is partially peeling off on the surface, or if the flaw width is generally 10 to 20 mm or less, depending on the steel type, the tip of the flaw is likely to be a block. After rolling, it is oxidized before reaching the first stand of the rough rolling mill of the hot strip mill, and a high-pressure water scale breaker (R8B, water pressure 100 kg/c) is installed at the entrance of the first stand.
77f or higher).

Therefore, if flaw cleaning is automatically performed based on surface information that includes noise signals from the above-mentioned secondary scale and flaw signals that do not become harmful defects in downstream processes, the noise signals may be mistakenly recognized as flaw signals. You will have to clean up the scratches, and you will have to clean up small scratches that don't need to be cleaned up, which will cause you to run and run, and your work efficiency will drop significantly.

Therefore, the present inventors scanned the surface of various hot steel materials within 30 seconds after cleaning, and determined the size distribution of deposits that gave a signal at a lower level than the sound part, where secondary scale is the main cause. Similarly, hot steel surface flaws (scratching, kamiko flaws, tongs, etc.) that give a lower flaw signal than the sound part
The size distribution of marks, etc.) was statistically analyzed.

FIG. 1 is a diagram showing a comparison of the results, with the frequency on the vertical axis and the length of deposits and flaws measured in the main scanning direction of the heated steel material on the horizontal axis.

From FIG. 1, it was found that the size of each surface flaw and the size of the secondary scale were different.

The purpose of the present invention is to focus on the above-mentioned points and to efficiently generate a flaw signal with an extremely good S/N ratio by appropriately processing surface information obtained from a scanning flaw detector based on the length in the main scanning direction. It is an object of the present invention to provide a method for detecting surface flaws in heated steel materials.

The method for detecting surface flaws in heated steel materials according to the present invention uses a scanning type flaw detector that has a flaw detection resolution capable of detecting flaws with a minimum dimension that should be detected in a small number of cases. is scanned to obtain an electric signal representing the surface condition of the hot steel material, and this electric signal is passed through circuit means for removing low frequency components and high frequency components to eliminate low frequency background signals caused by temperature fluctuations on the surface of the hot steel material. The present invention is characterized in that desired flaw information is obtained by removing high-frequency noise signals caused by foreign matter adhering to the surface of the hot steel material and micro-flaw signals smaller than the minimum critical dimension.

The present invention will be described in detail below with reference to the drawings.

FIG. 2 is a block diagram showing the construction of a hot flaw detection/care device that implements the method for detecting surface flaws in hot steel materials according to the present invention.

The apparatus shown in this example uses a scanning type flaw detector 1 to scan the surface of a hot steel material, and converts an electric signal representing the surface condition of the hot steel material into a low frequency removal circuit 2, a high frequency removal circuit 3, a signal amplification circuit 4, and a flaw detection circuit. The flaw information is input to a computer 6 via a discrimination circuit 5, edited by the computer 6, and then transmitted in real time to a flaw care device 7 for online flaw care.

For example, an infrared detector is used as the scanning flaw detector 1,
Scan the surface of the heated steel material in the width direction.

The electrical signal obtained here has a background that fluctuates due to the temperature drop at the edge of the hot steel surface and large temperature unevenness in the width direction, as shown by a in the figure.
Among the electric signals, there is a flaw signal F1. F2. In addition to F3, a high frequency steep noise signal N1. N2. Contains N3.

If such an electrical signal is passed through the low frequency removal circuit 2, a signal waveform with no background fluctuation can be obtained as already shown in the figure.If this is further passed through the high frequency removal circuit 3, a signal waveform with no background fluctuation can be obtained as shown in the figure.
As shown in , steep noise signals N□, N2. Flaw signal F1 from which N3 has been removed. F2. A signal waveform of only F3 is obtained.

This signal waveform is suitably amplified by the signal amplification circuit 4 and then input to the flaw discrimination circuit 5.

The flaw discrimination circuit 5 slices the amplified signal waveform at an appropriate level to obtain information about the degree of the flaw, and also obtains information about the size of the flaw by determining the width of the flaw signal waveform, and sends this information to the computer. Output to 6.

In addition to the information regarding the extent and size of these flaws, the computer 6 also contains signals regarding the type of flaw and the address of the flaw on the surface of the hot steel material, and provides appropriate information units (
For example, input data is configured in a format in which several words are serialized at 16 bits/word.

After the computer 6 edits the flaw information, it transmits it in real time to the control section of the flaw care device 7 to perform flaw care online.

In FIG. 2, a case will be explained in which the width of the defect to be detected and removed on the surface of the hot steel material is set to be more than 107W.

The flaw detection resolution of the scanning flaw detector 1 under these conditions is 1. Omrrt is enough, but
Of course, it may be less than that.

The electrical signal obtained from such a scanning flaw detector 1 is processed by a low frequency removal circuit 2, and then a high frequency noise component is removed by a high frequency removal circuit 3.

Here, as the scanning type flaw detector 1, the number of mirror rotations is 60.
rpS s If a configuration with 26 mirror surfaces and an effective viewing angle of 45° is used, and the width of the hot steel material is 2 doors, the passband frequency of the high frequency removal circuit 3 is O (i.e. D
-C) to 100KI (z), it is possible to effectively cut signals from surface-attached foreign matter and defects having a width of 10771.7J or less in the scanning direction of the hot steel surface.

Therefore, by using the device shown in Figure 2, noise signals from adhering foreign matter such as secondary scale that are inevitably included in surface information, and flaw signals from defects small enough to not become harmful defects in downstream processes can be effectively cut out. As a result, it is possible to efficiently detect and take care of only surface flaws that are actually harmful defects in the downstream process, so direct rolling can be easily realized.

As mentioned above, in the method of the present invention, the electrical signals from the scanning type flaw detector are appropriately processed to extract only the flaw signals that will become harmful defects in the downstream process. It is possible to detect defects with a high S/N ratio and efficiency even when they are present, and when cleaning defects based on this detection signal, it is possible to detect defects that are small enough to prevent them from becoming harmful defects in downstream processes. Since there is no need to remove it, the work efficiency can be significantly improved.

Furthermore, the scanning type flaw detector is extremely advantageous in terms of accuracy because it only needs to detect a small number of defects large enough to cause common damage.

The present invention is not only effective for detecting surface flaws on high-temperature slabs, billets, blooms, and beam blanks, but can also be effectively applied to detecting surface flaws on input hot coils, sheet bars, and the like.

[Brief explanation of drawings]

Fig. 1 is a diagram comparing the size distribution of the secondary scale and the size distribution of hot steel surface flaws, and Fig. 2 is a diagram showing a comparison between the size distribution of secondary scale and the size distribution of hot steel surface flaws. It is a block diagram showing the configuration of the mountain inspection and maintenance device. 1... Scanning flaw detector, 2... Low frequency removal circuit, 3... High frequency removal circuit, 4...
...Signal amplification circuit, 5...Flaw discrimination circuit, 6...
... Computer, 7 ... Scratch care device.

Claims (1)

[Claims] 1. When hot detecting surface flaws on hot steel materials, the surface condition of the hot steel materials is determined by scanning the surface of the hot steel materials using a scanning flaw detector that has a flaw detection resolution that can detect flaws with the minimum size that should be detected. This electrical signal is passed through a circuit that removes low-frequency components and high-frequency components to remove low-frequency background signals caused by temperature unevenness on the surface of the heated steel material, and to eliminate foreign matter adhering to the surface of the heated steel material. 1. A method for detecting surface flaws on hot steel materials, characterized in that desired flaw information is obtained by removing caused high-frequency noise signals and minute flaw signals smaller than a minimum critical dimension.