JPS601552A - Ultrasonic flaw detector - Google Patents

Abstract

PURPOSE:To restrain noise per wave transmitting-receiving sequence and make it possible to detect flaws at a high speed and a high sensitivity with a simple arrangement by outputting an ultrasonic echo detecting signal only when the amplitude is higher than the threshold value and further the signal width exceeds a threshold value. CONSTITUTION:In relation to an ultrasonic echo by a gate circuit 1 and a detecting circuit 2, any portion of the echo which has an amplitude exceeding a threshold value set in an amplitude threshold value setting circuit 4 is clamped in a clamping circuit 3. A signal width of the thus clamped signal is detected in a signal width detecting circuit 6, which value is compared in a comparison circuit 7 with a threshold value set in a signal width threshold value setting circuit 8. If not lower than the threshold value, an output controlling circuit 9 is controlled according to the control signal outputted in the circuit 7. Thereafter, noise of a flaw detection signal outputted via a signal processing circuit 5 and the circuit 9 is restrained by the controlling operation which is based on the results of distinguishing between the reflection echo and the noise waveform. Since noise is restrained every wave transmitting-receiving sequence, counting or other similar operations in accordance with sequential detections become unnecessary, thereby to effect the ultrasonic flaw detection at a high speed and a high sensitivity with a simple arrangement.

Description

[Detailed description of the invention] This invention injects an ultrasonic pulse into a material to be inspected.

The present invention relates to an ultrasonic flaw detection device that performs nondestructive flaw detection by detecting ultrasonic pulses (hereinafter referred to as nine reflected echoes) reflected from acoustically discontinuous surfaces such as defects in a material to be inspected. More specifically, it is equipped with a noise suppression circuit that reduces the influence of noise that is mixed in the time domain in which the reflected echo is predicted to be obtained and that is difficult to remove by applying a time gate to the flaw detection signal. The present invention provides an ultrasonic flaw detection device.

Conventional ultrasonic flaw detection equipment uses reflected echoes necessary for flaw detection and other noise signals9, such as continuous vibrations or multiple reflections of reflected ultrasonic pulses reflected by reflection sources that exist outside the flaw detection area, or the installation of flaw detection equipment. In order to distinguish between electromagnetic noise caused by the switching of other power equipment, depending on the location where the test is performed, an appropriate threshold is set for the amplitude of the flaw detection signal, and only the signals that exceed the threshold are extracted and processed. The following method is used.

FIG. 1 is a block diagram showing an example of a noise suppression circuit using this method. In FIG. 99, (1) is a gate circuit, (2I is a detection circuit, and +31rri clamp circuit.

(4) is an amplitude threshold setting circuit, and (5) is a signal processing circuit.

To briefly explain the operation, the flaw detection signal is gate circuit +11
gated in time, and only signals in the time domain corresponding to the desired flaw detection area are selectively extracted and input to the detection circuit (2). Detection circuit (21 is a circuit that performs half-wave rectification, full-wave rectification, or envelope detection on the input signal, and converts it into a waveform that is easy to perform signal processing in the signal processing circuit (51) described later.Detection The output of the circuit (2) is input to the clamp circuit +31, where it is softened by the threshold set in advance by the amplitude threshold setting circuit (4), and only the signal exceeding the threshold 1 is sent to the signal processing circuit (5).
1 is input. That is, the output of the clamp circuit (3) is
This circuit outputs input signals that exceed a set threshold value as is or amplifies them, and does not output signals that are smaller than the set threshold value. The signal processing circuit (5) is a circuit that converts the output of the clamp circuit 13+ into flaw detection data, and is a circuit for displaying the flaw detection signal on a CRT.

The amplitude of the reflected echo (hereinafter referred to as the reflected echo tube) and the time required for the reflected echo to be obtained after transmitting the ultrasonic pulse (hereinafter referred to as the reflected echo depth) are required as easy data for detection. The threshold value set by the O amplitude threshold setting circuit (4), which is the measurement circuit, is determined by checking the reflected echo level and noise level in the flaw detection signal in advance using a test piece whose shape and size of the defect are known. Based on the results, the reflected echo level is generally set to be small and higher than the noise level.

Only the desired reflected echo is output from the output of the clamp circuit (31).

However, conventional noise suppression circuits have the disadvantage that when noise larger than a set amplitude threshold is mixed in, the noise is mistakenly output as a reflected echo.

Examples of such noise include 1.For example, air bubbles enter the coupling material (usually a fluid such as water or oil) that acoustically couples the probe and the test material, and the ultrasonic waves are reflected by the air bubbles. The output appears in the flaw detection area due to multiple reflections or interference with other reflected ultrasonic waves, or electromagnetic waves generated by electrical switching jump into the flaw detection cable. There are cases where the output happens to appear within the flaw detection area. Although the influence of such noise on the flaw detection results varies depending on the flaw detection conditions or the flaw detection environment, generally flaw detection data is processed by a computer or the like.

This is a very serious problem in so-called automatic flaw detection equipment that determines the presence or absence of defects. In other words, this type of noise occurs asynchronously with the repetition rate of ultrasonic wave transmission, so even in the case of so-called manual flaw detection, in which skilled inspectors perform flaw detection without looking at a CRT monitor, etc. Even if noise is mixed into the flaw detection signal, it is easy to distinguish between the noise and the reflected echo just by looking at the way the signal appears, but in the case of automatic flaw detection equipment, it is difficult to set up such a judgment PA structure. It is from.

In some cases, a noise suppression method called a countdown method is adopted as one method for solving such problems. The countdown method is a method used to continuously detect flaws in the material to be inspected.The sending beam has a finite width, and the defects in the material to be inspected are also finite in size. This method takes advantage of the property that the height of one reflected echo changes continuously when flaws are detected continuously and finely. Figure 2 is a diagram showing the principle of the countdown method. This is a diagram schematically showing the reflected echo height E at that time and the noise N that happened to be mixed into the flaw detection area. The average value of the reflected echo E obtained in the wave transmission/reception sequence is output every five wave transmission/reception sequences.

As is clear from (aJ (1)J in Figure 2), even if a single noise is mixed in, the actual noise level as the average value of the five reflected echo heights will be suppressed to 4. The reflected echo of the defect F varies depending on the moving speed of the ultrasound probe P, the ultrasound beam transmission repetition frequency, and the size of the defect F.
Since it is normal to obtain up to three consecutive reflected echo heights, it is possible to obtain a considerably large average value output for five times. The cow sit down method takes advantage of this property, and stores the reflected echo heights obtained in each wave transmission/reception sequence in a memory circuit one after another in a memory circuit. It outputs the average value of the stored data.

In FIG. 2, the number of times the wave transmission/reception sequence is averaged is five times, but the more times the average is performed, the more sporadic noise is removed. However, as the average number of averages is increased, the average value of the reflected echoes also decreases, so it should be set taking into account the moving speed of the ultrasound probe, the ultrasound transmission repetition frequency, and the size of the minimum defect to be detected. It is something that should be done.

As described above, the countdown method is quite effective in suppressing noise, but it also has the following drawbacks.

(1) As shown in FIG. 2 (aJ and (bJ), the position where the average reflected echo quality is obtained is different from the actual defect position, so the position of the defect becomes unclear.

(2) If the defect is minute, the average reflected echo height will be relatively small and it will not be detected.

(3) The drawbacks of the above (llv 12+) become more pronounced as the ultrasound probe moves faster.

This invention was made in order to improve such drawbacks. 1. It detects the difference in waveform between reflected echo and noise, distinguishes noise, and outputs the flaw detection data for flaw detection signals that are determined to be noise. By not doing so, noise is suppressed. Therefore, noise determination according to the present invention is performed for each wave transmission/reception sequence, so there is no need to perform flaw detection continuously as in the countdown method, and since the ultrasonic probe does not have to be moved at high speed, flaw detection is possible. There is a first point in which the defect detection sensitivity does not decrease even when performing the above steps.

The operation of the noise suppression circuit of the ultrasonic flaw detection apparatus according to the present invention will be described in detail below with reference to FIGS. 3, 4, and 5. FIG. 3 is an example of a block diagram of a noise suppression circuit according to the present invention, and FIG.
~ (1)] When noise N is input [Figure 4 (a) ~
FIG. 3 is a diagram schematically showing waveforms in the main parts of FIG. Further, FIG. 5 shows an example of the configuration of a signal width detection section which is a main component of the noise suppression circuit according to the present invention.

In FIG. 3, (1) is a gate circuit, (2) is a detection circuit, (3) is a clamp circuit, and (4) is an amplitude threshold setting circuit.

(6) is a signal processing circuit. The basic operations of these circuits are the same as those shown in FIG. 1, but the purposes of operation of the clamp circuit (3) and amplitude threshold setting circuit (4) in FIG. 3 are different from those in FIG. That is, while the purpose of the clamp circuit (3) and amplitude threshold setting circuit (4) in FIG. 1 was to mask signals smaller than the threshold set by the amplitude threshold setting circuit (4), The purpose of those in FIG. 3 is to detect the point at which the input signal crosses a set threshold. In Fig. 3 (61 is the clamp circuit (output of 31,
In other words, it is a signal width detection circuit that detects the time width exceeding the set threshold of the input signal from one point of crossover between the input signal and the set threshold, and outputs that value in a digital or analog manner, (7) Compares the output of the signal width detection circuit (6) with the signal width threshold set in advance by the signal width threshold setting circuit (8) 1-9 Signal width detection circuit (6)
) is larger, the comparison circuit outputs a control signal to the output control circuit t9t. The output control circuit (9) temporarily stores the flaw detection data signal-processed by the signal processing circuit (51), and outputs it to a subsequent processing system (not shown) according to the control signal from the comparison circuit (7). The purpose is to output (no control signal) or output (with control signal).

Next, the operation will be explained using FIG. 4. Figure 4 (a
J~(7) is a schematic diagram when a reflected echo E is input, and FIG. 4(A)~(a) is a schematic diagram when noise N is input. As shown in Figure 4, noise due to multiple reflections and interference due to bubbles in the coupling material, electrical switching noise, etc. have a longer vibration duration than echoes reflected from defects. They are usually long. Therefore, their detection output (
For example, in the case of envelope detection), the results are as shown in Figure 4 (b) and Figure 4 (mouth), and the signal width exceeding the amplitude threshold Q (indicated by the horizontal broken line) is also compared with the case of reflected echo. Therefore, as shown in Figure 4 (C) and Figure 4 H, noise and reflected echo can be distinguished by focusing only on the signal width above the amplitude threshold. The Franz ° circuit (32 in Figure 3) outputs an output corresponding to the pulse in Figure 4 (C) and Figure 4 Hi3, and the signal width detection circuit (61
The width is measured by the signal width detection circuit, but as long as the signal width exceeding the amplitude threshold can be detected and measured, the output of the frang circuit 13+ can be of any shape. Any method may be used.

Signal width threshold RI for distinguishing between reflected echo and noise
l'i can be obtained by measuring the signal width of the reflected echo in advance using a test piece or the like. Then, by inputting this value to the comparison circuit (7) through the signal width threshold setting circuit (8), the signal is screened according to the signal width. As shown in the figure), a control signal is output to prevent the output of the flaw detection data temporarily stored in the output control circuit (9).

Therefore, since only the flaw detection data of the flaw detection signal whose signal width is narrower than the signal width threshold is always outputted from the output control circuit (9), flaw detection data mixed with noise is not outputted.

FIG. 5 shows an example of the configuration of the signal width detection section. In FIG. 9, Orj is an analog comparator, aυ.

09 is a D/A converter that converts external threshold data sd into a voltage value, and Q3 is an A/D converter that converts a flaw detection signal into flaw detection data.

(131 is an integrating circuit for integrating the output pulse of the analog comparator α and outputting a voltage proportional to the pulse width of the output pulse, including an op amp (13a) and a resistor (13
b) OI having a capacitor (13c) is an analog comparator of ξ when comparing the output voltage of the integrating circuit 0 and the output voltage of the D/A comparator, J (LS).

αe is a latch circuit for temporarily storing the output of the A/D converter tIz, Re is a reset signal, T is a data setting timing, and aS is a control signal.

Next, to simply explain the operation, the detected flaw detection signal is input to the A/D converter t1z, which is a signal processing circuit, and is also input to the analog comparator 4, and the amplitude threshold set by the D/A converter an A crossover point is detected. The output of the analog comparator 0I is a pulse corresponding to one crossover point as shown in Fig. 4 (C) and Fig. 4?M, so the integrator circuit (13
The output of 1 is the amplification factor of the integrator circuit, FiQ.
is the amplitude of the input pulse, and Δt is the pulse width) is output.

The output of the integrating circuit +13 is input to another analog comparator I.

There, it is compared with the output of the D/A converter RO 51 that determines the signal width threshold, and the analog comparator I detects one point of crossover between the output of the integrating circuit (131) and the signal width threshold, and detects a point corresponding to the crossover point. On the other hand, the output of the A/D converter (1'A) is input as data to the latch circuit αe and latched by the latch signal LS, but at that time, the output of the analog converter 1a4I, that is, the latch signal is Therefore, when there is no control signal, the output of the latch circuit αe is output from the A/D converter [12] directly as flaw detection data, and when the control signal is issued, it was latched to the latch circuit αe last time. The data is output as flaw detection data.

In the configuration example shown in FIG. 5, a voltage proportional to the signal width is generated to detect the signal width, but it is also possible to directly count the output pulses of the analog comparator (IQ) using the clock. Although the control signal is used to control the latch signal of the latch circuit Q61, it can also be used to control the output of the latch circuit (Ie).

Although the above description has been made regarding the case of a noise suppression circuit in an ultrasonic flaw detection device, the present invention is not limited to this, and can be applied to noise suppression circuits in general measuring instruments that use ultrasonic waves, radio waves, and optical pulse signals. good.

As described above, according to the present invention, the wave transmission/reception sequence 1
1g! Since reflected echoes and noise can be distinguished from each other for each time, there is an advantage that even in high-speed flaw detection, the influence of noise can be effectively reduced without reducing the detection sensitivity of minute defects.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an example of a conventional noise suppression circuit, Fig. 2 is a diagram explaining the operation of the countdown method used during continuous flaw detection, and Fig. 3 is a noise suppression circuit according to the present invention. FIG. 4 is a diagram schematically showing waveforms in the main parts of FIG. 3 when a reflected echo is input and when noise is input, respectively. FIG. 5 is a diagram showing an example of the configuration of a signal width detection section which is a main component of the noise suppression circuit according to the present invention. (1) is a gate circuit, (2: is a detection circuit, (3) is a clamp circuit, (4) is an amplitude threshold setting circuit, (5) is a signal processing circuit. (6) is a signal width detection circuit, (7) is a comparison circuit, (8) is a signal width threshold setting circuit, (9) is an output control circuit, 01 is an analog comparator, aυ is a D/A converter,
UZ is A/D converter, α3 is integration circuit, a4J is analog comparator, α9 is D/A converter, aQ
is a latch circuit. In the drawings, the same or corresponding parts are designated by the same reference numerals.

Claims (1)

[Claims] An ultrasonic flaw detection device that non-destructively detects defects in a test material includes a circuit that sets an amplitude threshold of a flaw detection signal for a defect in a test material, and detects a point where the flaw detection signal crosses the amplitude threshold. a signal width measurement circuit that measures a signal width equal to or greater than an amplitude threshold; a circuit that sets the signal width threshold; and a comparison between the output of the signal width measurement circuit and the signal width threshold;
A comparison circuit that outputs a control signal when the output of the signal width measurement circuit is larger than the signal width threshold, and an output of the first comparison circuit control whether or not to output the flaw detection data obtained from the flaw detection signal to the outside. An ultrasonic flaw detection device characterized by comprising an output control circuit that performs the following steps.