JP3006634B2 - Defect detection method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for detecting defects in an object to be measured having portions having different average thicknesses.
The present invention relates to a method and an apparatus for easily detecting which thickness portion has a defect.

[0002]

2. Description of the Related Art For example, automobile parts such as cylinder parts and piston parts used in piston mechanisms of engines are often made of cast metal. However, if these parts have defects such as cracks, cavities, dents, etc., these parts are destroyed. Can cause serious danger. Therefore, it is preferable that components having defects such as cracks, cavities, and dents can be detected and removed in a component manufacturing line before assembling an automobile. And if you can evaluate the location of the defect,
It also helps determine the cause of the defect.

Conventionally, methods for detecting and evaluating the position of a defect include a method based on the reflection of ultrasonic waves, a method based on sound when cracks are generated by AE (acoustic emission), an observation method using a CCD camera, and an X-ray photography method. And a color check method.

[0004]

Each of the above-mentioned conventional defect detection methods has the following handling problems.

[0005] That is, for example, the ultrasonic flaw detection method is a method of detecting a defect by bringing it into contact with an object to be measured and estimating a position thereof. The reflection due to the mismatch in the above and the waveform seen with a slight angle difference are different, and it is not easy to determine.

In addition, in the case of the AE method, the contact method is used as in the ultrasonic flaw detection method, and the measurement cannot be performed unless the crack is a progressive crack. Conversely, in the case of a progressive one, the measurement is performed while the crack is enlarged.

In addition, the observation method using a CCD camera has a drawback that the determination is disturbed even if there is a spot or a pattern other than a defect such as a crack or a dent, and a cavity called a “cavity” is detected in a casting or the like. And position cannot be determined.

[0008] Further, X-ray photography is effective because it can be visually observed directly, but adjustment of the X-ray dose is troublesome and observation is not possible.

As shown in FIG. 2, for example, in the cylinder part, irregularities 1 for retaining the engine housing part are formed on a part 3 of the outer peripheral part other than the joint part 2 of the mold. Portion 3 has a reduced average thickness. On the other hand, the seam portion 2 of the mold has a large thickness. Since the processed cylinder part has a thin part 3 and a thick part 2 with a small average thickness, if a defect occurs in this cylinder part, it is reduced to a part 3 with a small average thickness.
Even if it is a fatal defect if it occurs, if it occurs in the portion 2 having a large average thickness, it may be sufficiently usable as a part.

Therefore, if it is known whether the position where the defect occurs is the portion 3 having a small average thickness or the portion 2 having a large average thickness, it is important to determine whether the cylinder part is good or defective. Element. However, in the above-described conventional defect detection method, for that purpose, defect detection must be performed for each portion having a different thickness, which takes time and labor, is troublesome, and requires 100% inspection of the parts to be measured. Not suitable for inspection on a production line.

SUMMARY OF THE INVENTION In view of the above, the present invention provides a defect detection method which can easily detect which thickness portion of an object to be measured having portions having different average thicknesses has a defect and which is easy to handle. And a device.

[0012]

Means for Solving the Problems] defect detection method according to the invention, the vibration to be measured was added having a portion average thickness is different, and spectral analysis to pick up the vibration, the portion where the average thickness is different Of the spectrum corresponding to
Each of the different thicknesses according to each value of the average thickness
That the frequency position near detects by searching, the
Detected as corresponding to areas with different average thickness
Predefined frequency for each of the spectra
By detecting whether or not there are two peaks in the range, the presence or absence of a defect in each of the average thickness portions corresponding to the spectrum is detected.

Further, the defect detecting apparatus according to the present invention includes a vibrating means for applying vibration to an object to be measured having portions having different average thicknesses, and a pickup means for picking up the vibration of the object to be measured and converting the vibration into an electric signal. Receiving the electric signal from the pickup means, analyzing the spectrum of the vibration of the object to be measured, and performing scanning corresponding to the portions having different average thicknesses .
Each of the vectors to each value of the average thickness
Search by searching near different frequency positions
Means for output, also the average thickness corresponds to a different portion
For each of the spectra detected as
Has two peaks within the specified frequency range
Corresponding to the spectrum by detecting whether
Means for detecting the presence or absence of a defect in each of the average thickness portions to be formed .

[0014]

The vibration of the measured object can be picked up in a non-contact manner. As a result of research by the inventors of the present invention, when spectral analysis the picked-up vibrations in the object to be measured, each portion having an average thickness differs
Different frequency locations determined in accordance with the, if for example the cylinder component, the frequency position correspondingly different average thicknesses of the thin portion 3 and average the thick portion 2 thicknesses each one-to-one, the spectrum appears. Then, it has been found that when there is a defect in each of the thickness portions 2 or 3, it can be observed that each corresponding spectrum is separated into two.

This can be considered as follows.

That is, due to the vibration, the portion 2 having a small average thickness and the portion 3 having a large average thickness generate a unique vibration determined according to the average thickness (substantially mass), and the vibration of the vibration is generated. Therefore, a spectrum standing at a frequency position corresponding to 1: 1 in each average thickness portion can be observed. When there is no defect, the spectrum can be observed as having one peak.

However, if there is a defect, the vibration cannot be transmitted to the defective portion, so that a bypass propagation path is generated, and the energy of the vibration is divided into the energy of the inherent vibration and the energy of the vibration passing through the bypass. Since the vibration passing through the detour has a longer propagation path than the natural vibration, the frequency at which the spectrum rises is lower than the spectrum due to the natural vibration, and therefore, the spectrum is divided into two.

In this manner, whether or not the thickness portion of the measured object has a defect depends on whether the spectrum of the natural vibration corresponding to each of the different average thickness portions can be separately observed. Can be detected.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings, taking as an example the case of detecting a defect in the above-mentioned cylinder part shown in FIG.

In this example, vibration is applied by giving an impact to the part to be measured. Then, the vibration is picked up by a displacement meter or a vibration detection sensor having a sharp directivity, and the natural vibration is subjected to spectrum analysis. Then, the part to be measured is 1
With respect to the vibrations of the following orders, second,..., A frequency spectrum having peaks at frequency positions corresponding to the portion 2 having a large average thickness and the portion 3 having a small average thickness is obtained. The frequency at which the spectrum has a peak is determined by the shape, material, and size of the measured object.

FIG. 1A shows the spectrum of the primary vibration when no defect exists in the cylinder part.

That is, the vibration causes the cylindrical cylinder part to vibrate in a first order, a second order,..., And vibrates in each order, and the frequency of the natural vibration depends on the thick part 2 and the thin part 3 of the cylinder part. Frequency f determined according to the average thickness of
_A, it has become a f _B. In this case, the frequency of the vibration of the portion 2 having a large average thickness is relatively high, and the frequency of the vibration of the portion 3 having a small average thickness is relatively low.

Therefore, as shown in FIG. 1A, the peak 11 of the spectrum at the high frequency position corresponds to the portion 2 having a large average thickness, and the peak 12 of the spectrum at the low frequency position corresponds to the portion 3 having a small average thickness. I do.

FIG. 1A shows a case where there is no defect in a cylinder part which is a part to be measured. On the other hand, when there is a crack (hereinafter referred to as a crack) penetrating the wall surface of the cylinder of the object to be measured, When attention is paid to an odd-order spectrum such as a first-order or third-order spectrum, the peak of the spectrum corresponding to the thickness portion 2 or 3 where the crack exists can be separately observed. That is, if a crack exists at any position of the cylinder part, for example, a first-order spectrum corresponding to the thickness portion 2 or 3 where the crack exists is located at the frequency position of the peak 11 or 12 of the spectrum, and Separate from the lower frequency position, 2
It can be observed to have two spectral peaks. The reason for this can be explained as follows.

For example, when a crack (crack) is present as a defect, an oscillating wave cannot pass through the crack, and an oscillating wave bypassing the crack occurs. For this reason, the vibration energy of the average thickness portion where the crack exists is divided into the energy of the vibration wave unique to each of the portions 2 and 3 and the energy of the vibration wave bypassing the crack. Since the propagation path of the vibration bypassing the crack is longer than that in the case where there is no crack, the spectrum due to the vibration wave generated by this crack appears at a position lower in frequency by that much, and as a result, the crack becomes The spectrum corresponding to the existing thickness portion is observed separately in two parts. Note that the sum of the energies of the two separate spectra is equal to the energy of the corresponding primary spectrum in the case where there is no crack in FIG. 1A.

FIG. 1B shows a primary spectrum in a case where a crack is present in a portion 3 having a small average thickness, and 13 shows a spectrum that appears due to the crack. The size of the crack is proportional to the frequency difference Δf at a frequency position where two separate spectra stand, as shown in FIG. 1B. Here, the size of a crack refers to its volume, but since the width of a crack is so small as to be almost negligible, it substantially indicates the length of the crack. In the case of the cylinder part of this example, it was confirmed that the frequency difference Δf = 5 Hz indicates the presence of a crack having a length of 4 mm.

Next, if there is a non-penetrating defect such as a cavity or a dent in the cylinder part of the object to be measured which does not penetrate the wall surface, if there is no other crack which is a penetrating defect,
The odd-order spectrum such as the second or third order spectrum is not divided into two, and non-penetrating defects such as cavities cannot be detected only by paying attention to the spectrum of the odd-order only. This is because the vibration that appears as the primary spectrum is a vibration along the circumference, and if it does not penetrate the cylindrical wall, such as a dent, it does not require a detour and does not split into two peaks. is there.

However, if attention is paid to the even-order spectrum, for example, the second-order spectrum, when a portion such as a dent is viewed in the thickness direction, a detour path can also be considered. It can be observed that the spectrum corresponding to each thickness portion where occurs is split into two.
Also in this case, the frequency difference between the two separate spectra is proportional to the size of the non-penetrating defect.

Therefore, these non-penetrating defects can be detected in exactly the same manner as cracks.

When the defect is minute, the Q value of the spectrum corresponding to each of the thickness portions 2 and 3 (= (f ₁ −
f ₂ ) / f ₀ (see FIG. 3) increases, and the width increases. This is considered to be because the fundamental natural vibration spectrum and the spectrum of the vibration due to the defect are observed as being combined without being separated due to the frequency resolution of the arithmetic unit.

Therefore, in this case, the presence or absence of a defect can be determined by detecting the magnitude of the Q value of the spectrum corresponding to each of the thickness portions 2 and 3.

Note that the sum of the energy of the fundamental natural vibration corresponding to each of the thickness portions 2 and 3 and the energy of the vibration due to defects such as cracks, cavities, recesses, and the like is shown in FIG. As described above, it is equal to the energy of the spectrum corresponding to the bulging portion.

Next, an embodiment of an apparatus to which the above-described defect detection method is applied will be described with reference to FIG.

FIG. 4 shows an embodiment of the defect detection apparatus according to the present invention, which is an example in which a defect of the above-mentioned cylinder component as a component to be measured is detected.

The cylinder part as the object to be measured 21 is moved by the transfer device 23 controlled by a control device 22 having a microcomputer, for example, to the measurement stage 2.
4 and loaded.

The measurement stage 24 is made of, for example, hard rubber or foam. When the measurement object 24 is mounted on the measurement stage 24 by, for example, a sensor provided on the measurement stage 24, the control device 22 drives the vibration device 25 to The object 21 is vibrated. In this example, the vibration device 25 performs, for example, impulse impact on the DUT 21 with an impact object such as a pendulum-shaped weight. The drive mechanism of the weight is constituted by, for example, a cam mechanism so that the weight is immediately separated from the object to be measured after the impact.

When a position where a defect is present is just vibrated, the vibration due to the defect hardly occurs. Therefore, in order to reliably detect the defect, it is preferable to perform the vibration at at least two places. Further, in the case of a cylindrical part, at the position 180 degrees different from the vibration position, the vibration occurs in the same phase as the vibration position.
Since vibrations of opposite phases occur at positions different by 0 degrees, even if a position different by 90 degrees from one vibration position is vibrated, only the same result is obtained. Therefore, the two excitation positions are selected while avoiding the excitation positions that differ by (90 × n) degrees.

In this example, the vibration position is, for example, 22.
There are two locations that are angularly separated by 5 degrees. For this reason, in this example, the measurement stage 24 is rotatable in a horizontal plane, and is positioned and mounted so that the center line position of the cylinder component, which is the object to be measured 21, matches the rotation center position of the measurement stage 24. Is placed.

Then, first, the DUT 21 is vibrated by the vibrating device 25, and then the measuring stage 24 is
By rotating by 2.5 degrees, the DUT 21 is further vibrated.

The object 2 to be measured that has been vibrated as described above
The vibration of No. 1 is the non-contact sensor 2 of the output vibration receiving device 26.
7, the signal is converted into an electric signal, and predetermined signal processing is performed by the signal conditioner 28. Sensor 2
For 7, any device that can detect vibration can be used, and a displacement meter or the like can be used. However, in order to minimize noise vibrations from the surroundings, it is preferable to have a sharp directivity in the direction of the DUT 21.

The signal conditioner 28 amplifies the electric signal and removes unnecessary high and low frequency components (removal of a trend).

The electric signal from the output vibration receiving device 26 is
The data is supplied to the arithmetic processing / judgment device 30 via the transmission line 29. The arithmetic processing / judgment device 30 has, for example, a microcomputer, and performs arithmetic processing and determination operations to be described later by software. This processing is represented by functional blocks as shown in the figure.

The vibration to be considered here is the natural vibration of the shape of the measured object. However, when the object to be measured is forcibly vibrated, a longitudinal wave is initially generated similarly to the forced vibration or the seismic wave, and this is mixed with the natural vibration. If the defect is a considerably large crack or dent, the above method may be able to detect the defect even if other than these natural vibrations are present. However, it is usually difficult to detect a defect unless the natural vibrations are removed as much as possible.

Therefore, in this example, this is solved as follows.

That is, when the DUT 21 is vibrated,
There are a sine wave method and an impulse impact method. In the case of the sine wave method, the device under test 21 is vibrated under certain conditions and stopped at a certain moment. Then, the measurement of the vibration is started when a short time has passed since the stop.

In the case of the impulse impact method, the measurement is started at a point in time when a short time has passed immediately after the vibration is applied by applying an impact or the like.

In this case, the time from when the vibration is stopped or when the impact is started until the measurement is started can be determined as follows. That is, the speed c of the sound wave propagating through the object 21 differs depending on its Young's modulus E (elastic coefficient) and the density ρ of the object, and is obtained from the relationship of c = (E / ρ) ^1/2 .

For example, in the case of the impulse impact method of this example, since the cast iron part is the object 21 to be measured, the velocity of the longitudinal wave is 4560 m / s, and the transverse wave is 1 / 1.8, that is, about 2780 m / s. FIG. 5A shows a time-series waveform of the vibration picked up from FIG. In this waveform, a transverse wave is detected after a portion of only a fast longitudinal wave lasts about 26 μsec. Then, the vibration attenuates exponentially after the peak value of the vibration of the transverse wave, and the vibration gradually stops.

As can be seen from the waveform of FIG. 5A, the vibration after the excitation is the same as that of the seismic wave, so that vertical waves and high-speed waves are mixed as described above. , A forced vibration remains, and a natural vibration waveform peculiar to the shape of the DUT 21 is not obtained. It is considered that the natural vibration wave peculiar to this shape is observed shortly before stopping, for example, like a “dice motion” of a coma. Therefore,
In this case, the vibration after the peak value of the transverse wave and starting to attenuate is extracted. Therefore, to set the window W ₁ of the rectangular wave as shown in FIG 5B, this window W _1, in this example extracts the vibration wave.

That is, the electric signal input to the arithmetic processing / judgment processing device 30 is supplied to the gate means 31. Then, based on the above-mentioned window signal W ₁ from the window W ₁ forming means 32, the DUT 2 after the vibration, that is, the impact is applied.
From the first vibration, a natural vibration component of the shape of the DUT 21 is extracted. In this example, from the point of 20msec elapses from immediately after the impact launched window W _1, sets a window width of 200 msec. For this reason window W ₁ forming means 32, window W ₁ is formed based on the vibration start information from the control unit 22.

[0051] As described above, the natural frequency component of the profile of the workpiece 21 is extracted by the window W _1.

Then, the natural vibration portion is converted into digital data by the A / D conversion means 33 and written into the memory means 34. Then, the digital data is read from the memory means 34, and the waveform emphasis means 35
A window W ₂ forming means 3 is used for the digital data.
Emphasizing window W ₂ from the 6 is applied. The emphasis for the window W ₂ is as follows.

[0053] That is, even if extracted natural vibration waveform portion specific to the shape of the object to be measured 21 by window W _1, minute defects is likely hides the spectrum of the basic natural frequency, spectrum as described above The Q value of must be detected.

Therefore, it is conceivable to minimize the spread of the "tail" of the spectrum waveform of the fundamental natural vibration so as to make it easy to determine the presence or absence of a defect. For this purpose, it is sufficient to apply a correction such that the spectrum waveform as shown in FIG. 3 is rapidly attenuated (the Q value does not change) as shown by a broken line 19 in FIG. .
In this way, the "tail" of the spectrum waveform becomes narrow, and even if it is a minute crack or dent, the minute defect is not a Q value of the spectrum but a fundamental natural vibration spectrum and a spectrum of the vibration due to the defect. Can be separated and detected.

[0055] In order to emphasize the spectrum as described above, the picked-up vibration waveform, it further multiplied by emphasizing window W ₂ of the waveform having the following formula.

Y = acos ² (xωt) + bcos ² (xωt + τ) +... + Kcos ² (xωt + nτ) + C Here, τ indicates a time delay, for example, λ / 4 (λ is a wavelength). In this example, a = b =... = K. The emphasizing window W ₂ has a waveform as shown in FIG. 5C.

FIG. 6A shows the measured object 21 picked up.
Window for natural vibration extraction W ₁
And shows a spectrum of vibrations entire portion before applying emphasizing window W _2. Further, FIG. 6B shows the spectrum of the vibration extracted from after 20msec elapses from immediately after the impact of the vibration of the measuring object 21 by the natural frequency extraction window W _1, and the spectrum of the vibration due to the fundamental natural frequency spectrum and defect Separation can be observed. Further, FIG. 6C is a spectrum waveform after subjected to emphasizing window W ₂ mentioned above, it can be seen that the fundamental natural vibration spectrum, a vibration spectrum of defects such as cracks or dents may be more clearly separated.

[0058] window W _2, like the window W _1, the window W ₂ forming means 36, the control device 2
2 is formed based on the information of the start of the vibration from the second.

The data after being emphasized in this way is supplied to the spectrum analyzing means 37, where the spectrum is analyzed.

The defect presence / absence determining means 38 uses the spectra corresponding to the respective thickness portions 2 and 3 as shown in FIG. 1 obtained as a result of the spectrum analysis, for example, as follows. The presence or absence of a defect is determined by determining the existence of a spectrum of vibration due to a defect in the portions 2 and 3, and the position of which average thickness portion is defective is evaluated.

The defect presence / absence judging means 38 first obtains spectra corresponding to the portions 2 and 3 having different average thicknesses by searching near the frequency position at which the portions 2 and 3 should stand. For the corresponding spectrum, whether or not there is a defect in the thickness portion is determined based on whether or not the spectrum is divided into two.

That is, as shown in FIG. 7, from the spectrum waveform obtained by the spectrum analysis means 37, a predetermined thickness portion is detected to detect a spectrum corresponding to the thickness portion to be determined. in corresponding primary spectral frequency range d ₁ and the secondary spectrum in the frequency range d _2, obtains the peak value from those respective amplitudes large to turn for example, five, and stores the frequency and the peak value. Next, with respect to the primary and secondary spectra, the frequency ranges d ₃ and d ₄ (d ₃ and d ₄ <) in which the spectrum of the fundamental natural vibration of the thickness portion and the spectrum of the vibration due to the defect are considered to form a pair. d ₁ , d ₂ ) are determined in advance, and a search is made as to whether or not there is a pair of the five peak frequency values within the frequency ranges d ₃ , d ₄ .

When the pair of the primary spectrum is detected, the higher frequency of the lower frequency pair is recognized as the position of the primary fundamental natural vibration spectrum, and the frequency width is determined based on the frequency position. narrower than d _3, (which may be a peak of course pairs) decides whether or not there another peak from the basic natural oscillation spectrum in the frequency width d ₅ which is predetermined, if there is a peak, the thickness portion thereof Is determined to have a crack.

Similarly, when a pair is detected in the secondary spectrum, the higher frequency of the lower frequency pair is recognized as the position of the secondary fundamental natural vibration spectrum, and based on the frequency position, the determined whether another peak is present a principal natural vibration spectrum to a frequency width d ₆ that is narrow is predetermined from frequency width d _4, if there is a peak, nest or recess or the like in the thickness portion It is determined that there is a non-penetrating defect.

FIG. 8 shows a flowchart of the defect presence / absence determination operation of the defect determination means 38 for one thickness portion described above.

This operation for determining the presence or absence of a defect is performed on the spectra corresponding to the thick portion 2 and the thin portion 3 having the average thickness, respectively, and it is determined whether or not a defect exists in each of the thick portions 2 or 3.

When the defect judging means 38 judges that there is a defect, the size of the defect is obtained from the frequency difference between the two separate spectra.

The result of the presence / absence determination of each thickness portion by the defect determination means 38 and the information on the size of the defect are determined as non-defective products.
It is transmitted to the defective product judging means 39, and the non-defective / defective product judging means 39 judges whether the object to be measured is a non-defective product or a defective product based on the information on the position and size of the defect. In the case of this example, the threshold value for distinguishing a non-defective product from a non-defective product differs depending on the average thickness of a portion where a defect has occurred.

The information from the non-defective / defective product judging means 39 is transmitted to the control device 22, and the part to be measured judged as a defective product is removed from the line by the sorting device 40. For example, a marker attaching means (not shown) prints, using paint or the like, whether there is a defect in the thin part 3 or a defect in the thick part of the component to be measured (for example, a mark of a different color corresponding to the thickness). Can be added).

The above description is an example in which the object to be measured is a cylindrical cylinder part. However, the object to be measured is not limited to this, and the average thickness may be three or more different thickness portions. The present invention is applicable to a device under test having

The vibration method is not the impulse impact method, but various vibration methods can be employed, such as, for example, fixing one end and applying a bias to the other end to generate vibration.

[0072]

As described above, according to the present invention, an object to be measured having a plurality of portions having different average thicknesses is vibrated, and the respective average thicknesses of the object to be measured are non-contacted by a sensor. By detecting unique vibrations of different portions and performing spectrum analysis of the vibrations, it is possible to easily detect and determine whether or not a defect exists in each thickness portion. In other words, it is possible to easily evaluate the position of the thickness portion of the plurality of different thickness portions where the defect occurs.

Further, since the method and the apparatus for detecting the position of a defect by detecting the inherent vibration of the object to be measured in a non-contact manner, the sensor can be used as in the case where the sensor is in contact with the object to be measured. There is no irregular reflection due to mismatch at the time of contact,
The waveform is simple and easy to determine. That is, when detecting the presence or absence of a defect occurring in a plurality of different thickness portions of the measured object, when determining the position, there are few factors that disturb the determination,
Stable defect determination is possible.

According to the present invention, if the object to be measured can be distinguished from natural vibration even if it has wrinkles or irregularities,
There is a feature that a defect can be detected and its position can be evaluated.

Another feature is that the defect can be determined not only partially but also for all of a plurality of different thickness portions by vibrating the workpiece at least once at a position determined by its shape. is there.

[Brief description of the drawings]

FIG. 1 is a spectrum diagram for explaining a defect detection method according to the present invention.

FIG. 2 is a diagram for explaining an example of a component to be measured according to the present invention.

FIG. 3 is a diagram for explaining the present invention.

FIG. 4 is a block diagram of an embodiment of a defect detection device according to the present invention.

FIG. 5 is a diagram for explaining a window waveform for natural vibration extraction and emphasis.

FIG. 6 is a diagram for explaining natural vibration extraction and emphasis.

FIG. 7 is a diagram for explaining a defect determination method according to the present invention.

FIG. 8 is a flowchart of an example of the operation of the defect determination method according to the present invention.

[Explanation of symbols]

 2 thick part of average thickness 3 thin part of average thickness 11 spectrum corresponding to part 2 of large average thickness 12 spectrum corresponding to part 3 of small average thickness 13 spectrum due to defect 21 DUT 25 vibration device 26 Output Vibration Receiving Device 27 Vibration Receiving Sensor 30 Arithmetic Processing Judgment Device

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-3-120458 (JP, A) JP-A-61-73063 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01N 29/00-29/28

Claims (2)

(57) [Claims] 1. A vibration is applied to an object to be measured having a portion having a different average thickness, the vibration is picked up and spectrum analysis is performed, and a spectrum corresponding to the portion having a different average thickness is obtained.
Are different frequencies depending on the respective values of the average thickness.
It is detected by searching near the position, and is detected as corresponding to the portion where the average thickness is different.
For each of the measured spectra,
By detecting whether having two peaks within a wave number range, characterized by detecting the presence or absence of defects in each of the average thickness portion corresponding to the spectrum
Defect detection method to be. 2. Vibration means for applying vibration to an object to be measured having portions having different average thicknesses; pickup means for picking up vibration of the object to be measured and converting the vibration into an electric signal; Receiving a signal and analyzing the spectrum of the vibration of the object to be measured, wherein the average thickness is different
Each of the spectra corresponding to the portions is represented by the average thickness
Search near different frequency positions according to each value of
Means for detecting the difference in the average thickness.
For each of the measured spectra,
Detects whether there are two peaks within the wave number range
The average thickness corresponding to the spectrum
Defect detection apparatus according to claim Rukoto and means for detecting the presence or absence of defects in each minute.