JPS5920969B2 - Object surface flaw detection device - Google Patents

Description

[Detailed description of the invention] The present invention relates to an apparatus for detecting surface flaws on an object.

Traditionally, devices for detecting flaws on machined metal surfaces include an optical system component that shines light on the surface to be inspected and collects the reflected light, and a photoelectric converter of the focused light to obtain the results during preliminary measurements. Some devices include electrical system components that use an amount proportional to average brightness as a defect detection level to determine pass/fail during actual measurement. The outline of the above device is shown in Fig. 1. Preliminary measurements are performed by an analog gate 1, an integrating circuit 2, and a level adjuster 3, which are shown in broken line parentheses, and the output signal f of the level adjuster 3 is used as a defect detection level for a comparator. The output signal a from the photomultiplier tube 5 is compared with the output signal a from the photomultiplier tube 5. When inspecting a mixture of objects with different levels of surface polishing, this conventional device uses a The signal will be at various levels depending on the situation, but the defect detection level f will also move in proportion to the average brightness determined during preliminary measurement.
B, No matter what level the signal f is, the defective signal L
capture. Furthermore, even if the signal is entirely dark C, it can be determined as a good signal if there is no defective signal L, so that it can produce some results.
However, for example, in the case of a thin rust-like defect K appearing on a relatively wide area on a good surface J as shown in FIG. It comes out, but it doesn't come out in a very dark direction. Therefore, the average brightness of the entire surface of the object to be inspected during preliminary measurement is pushed up in the dark direction due to this type of defect, and as a result, the defect detection level f moves in a direction where it is difficult to capture the defect signal L. Therefore, if the defect detection level f approaches the signal for the purpose of detecting a defect with a small defect signal from the average level of the signal, even if it is a good product, the signal will reach the defect detection level f due to the fine structure of its surface. This may occur, resulting in the inconvenience of incorrectly determining that the product is defective. SUMMARY OF THE INVENTION An object of the present invention is to provide a surface flaw detection device for an object that eliminates the above-mentioned drawbacks and has high defect detection accuracy.

To explain an embodiment, as shown in FIGS. 5 and 6, the present invention includes a pusher 16 for transporting the object X from position N1 to position N2, and a roll 7 for rotating the object X at position N2. A discharge chute 9 having a sorting door 8 through which the inspected object X at position N2 is pushed out by the inspected object an accessory part consisting of;
As shown in FIG. 7, a light source 10 illuminating the object X is installed above the object X, a condenser lens 11 and a half mirror 12 are installed between the light source 10 and the object An imaging lens 13 that forms an image of the light reflected from the surface of the object X guided by the half mirror 12, and an eighth surface that removes excess light and forms a thin line image Q at the imaging position by the imaging lens 13. A fixed slit 14 as shown in FIG.
A rotary slit 15 rotated by a motor 16 and having a plurality of slits 15a radially and equally spaced from the axis center as shown in FIG. An optical system component 18 includes a condenser lens 17 that guides the light transmitted through the intersection of the slit 11 and the rotating slit 15 to a photomultiplier tube 20 of an electrical system component 19, which will be described later, and the above-mentioned optical system component 18 as shown in FIG. Optical system component part 1
The electrical system component 19 is installed after the optical system component 8 and consists of a main measuring device and a preliminary measuring device that converts the light from the optical system component 18 into an electrical signal and determines the quality of the surface of the object to be inspected. be done.

The measuring device includes a photomultiplier tube 20 that converts light from the optical system component 18 into an electrical signal, an amplifier 21 that amplifies the output signal A from the photomultiplier tube 20, and an amplifier 21 that amplifies the output signal A from the photomultiplier tube 20. A comparator 22 that compares the output signal of the width detector 21 with the defect detection level F from the preliminary measurement device and outputs a pass/fail judgment signal, a gate circuit 23 that passes the output signal G of the comparator 22 for the actual measurement time, and Output signal H of gate circuit 23
is held during the loading and unloading time until the inspected object It is composed of a memory circuit 24, and a relay circuit 25 which is turned on and off by the output signal 1 of the memory circuit 24, and sends a signal for moving the aforementioned sorting door 8 and sorting good products from defective products. The preliminary measurement device is disposed between the amplifier 21 and the comparator 22 as shown in the dashed line in FIG.
, 2, 3, and the preliminary measurement signal 1,
analog gates A, , a2 which are enabled to operate cyclically by 2 and 3 and which pass the output signal A amplified by the divided pre-measurement times T, , t2 and t3, respectively;
, a3 and the output signals Bl, B2, B of the analog gate
3, and the integral value is calculated at the preliminary measurement time Tl, t2.
, t3, the integration circuits B1, b2, b3 hold their values during the preliminary measurement signals 2, 3, 1 following the subsequent preliminary measurement signals 3, 1, 2, and are reset by the subsequent preliminary measurement signals 3, 1, 2, and each preliminary measurement signal 2.
,3,1, and the integrating circuit B,,b2,
Analog gates Dl, d2, d3 pass the outputs Cl, C2, C3 of b3, and analog gates Dl, d2, d3 pass through the analog gates Dl, d2, d3, which are made operational at the start of the preliminary measurement and reset at the completion of the measurement. a peak hold circuit 28 that holds the maximum value of the integral circuit outputs Dl, D2, and D3; and a level adjuster 29 that sets the defect detection level F to the relationship F-AE according to the degree of the defect for which the maximum value is to be detected. It consists of

Here, α is a constant set by the level adjuster 29, and is derived empirically from past accumulated data. The electrical system component 19 has a configuration in which the output signal of the level adjuster 29 is set as a defect detection level F and is compared with the output signal A from the photomultiplier tube 20 by a comparator 22. As shown in the figure, the output signal A from the photomultiplier tube 20 is once divided by the output signal E of the peak hold circuit 28 by the divider 30 to equalize it to the reference level, and then the comparator 22
A similar output signal can be obtained by comparing the defect detection level F' set to a constant value by /. That is, in the method shown in FIG. 10 described above, the output signal A
is f(t), and the maximum value E of the integral value is f(t)Ma
x, the ratio R between the signal A and the defect detection level F is, but in the method using a divider, the divider is f(t)
The signal passed through becomes:=::, and this signal has a lack of α, Flf.
(1)'111[j'L When the defect detection level is set, the ratio is as follows, which is equal to the above.

Therefore, whichever method is adopted has the same effect in detecting defective signals.

The quantity R here is a quantity indicating the degree of setting of the defect detection level. However, the oscillation frequency of the oscillator 26 is kept constant, and therefore each preliminary measurement signal 1, 2, 3 is kept at a constant time. Also, the integral constant of the integrating circuit is the output C of the integrating circuit.
l, C2, and C3 are adjusted so that they are each approximately equal to the average brightness of the surface that is the subject of the preliminary measurement. The operation of this device having the above configuration will be explained below.

One cycle in which the inspected object Loading and unloading time for dropping the roll 7 into the discharge chute 9 below the roll 7 after inspection, and the roll 7
The measurement time is divided into the measurement time during which surface flaws are detected by irradiating light onto the object X rotating above.

Therefore, the measurement time is divided into a preliminary measurement time T1 in the first stage and an actual measurement time T2 in the second stage, and further, the preliminary measurement time T1 in the first stage is divided into preliminary measurement time t1 t T2? T3 la t1? It is subdivided into T22゜゜゜゛゜゛゜゜゜゜゛゜゛゜゛.

At this time, it is desirable that the surface to be preliminarily measured and the surface to be actually measured are the same surface, but even if the whole is estimated by preliminary measurement of a part, it is practical if the preliminary measurement part has a sufficient range. There is no problem, and in this embodiment as well, one part is preliminarily measured, and the subsequent part is actually measured. In addition, the surface to be subjected to preliminary measurement is only a part of the surface at a certain preliminary measurement time, and as shown in FIG. 13, the entire width of the object to be inspected ,
For a specific inspected object X, depending on the characteristics of its processing method,
If defects often occur on a certain surface,
When intentionally avoiding that part, for example, the area inside the dotted line in FIG. 13 may be excluded from the preliminary measurement. Now, the reflected light from the surface of the object to be inspected X passes through the camera lens 12, the imaging lens 13, and the fixed slit 14, and a thin line image Q is projected onto the rotating slit 15.

The positional relationship between the thin line image Q and the rotating slit 15 is such that both ends of the thin line image Q can simultaneously touch both adjacent slits 15a, as shown in FIG. As the rotating slit 15 rotates, the thin line image Q is scanned and is incident on the photomultiplier tube 20 by the condensing lens 17. The output signal of the photomultiplier tube 20 is appropriately amplified by an amplifier 21, and a signal A as shown in FIG. 15 is sent to a preliminary measurement device. 15th
In the figure, M indicates a signal from a good surface, and L indicates a signal from a defective portion, which appears in the dark direction. R is a signal when the slit 15a of the rotating slit 15 is applied to the image at the same time as shown in FIG. 14, and it appears in the bright direction. This output signal A is first sent to an analog gate a1 enabled by a preliminary measurement signal 1, and the analog gate a1 sends an output signal B1 to an integrating circuit B for a preliminary measurement time T. The integrating circuit B integrates the output B1, holds its value during the preliminary measurement signal 2, and is reset by the preliminary measurement signal 3. The same applies to the integrating circuits B2 and b3, and this situation is as shown in FIG. The analog gate D, is enabled with the preliminary measurement signal 2 and leads the output C1 of the integrating circuit b1 to the peak hold circuit 28. Thus, the peak hold circuit 28 receives the output signals Dl, D2, D3 of the analog gates D, d2, d3 sent to the peak hold circuit 28.
The maximum value E is held L, that is, the brightest signal during preliminary measurement, in other words, the output signal value of the best part is held. The output E is increased or decreased by a level adjuster 29 to provide a defect detection level F of the comparator 22.
FIG. 17 shows the output F of the level adjuster 29 based on preliminary measurements.
The result obtained in the preliminary measurement 1 is obtained as the output F of the level adjuster 29 during the preliminary measurement 2 due to the circuit configuration. However, this is the case when the integral value is larger than the maximum value of the integral values found in the past, and when it is smaller, the past maximum value becomes the output F of the level adjuster 29 as it is. The same holds true for preliminary measurements 2 and 3. The right side of FIG. 17 shows a situation where an amount proportional to the maximum value of the integral value obtained during preliminary measurement is used as the defect detection level F, and even defects with relatively small defect signals can be detected.
FIG. 11 shows a part that achieves the same purpose as the level adjuster 29 and comparator 22, but in this case, the output signal A from the photomultiplier tube 20 is divided by the divider 30. is divided by the output signal E of the peak hold circuit 28 corresponding to the output signal A, the output signal is once shifted to the reference level, and then the comparator 22! This is compared with the defect detection level F' which is set to a constant level. Comparators 22 and 2
The output signal G of 2' passes through the gate circuit 23 that passes the input signal in synchronization with the actual measurement period, and is stored in the next memory circuit 24 during the aforementioned unloading period, and the output signal 1 of the memory circuit 24 causes the relay circuit 25 is activated. The power connected to the relay circuit 25 operates the sorting door 8 in response to the presence or absence of a defect signal. According to the present invention, it is possible to detect a defect with a relatively small defect signal based on the average level of the signal.

For non-defective products, the defect detection level does not become too close to the waveform, so that it is not mistakenly determined to be defective.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram of a conventional device, Fig. 2 is a diagram showing the relationship between the inspection signal and defect detection level in an embodiment of the conventional device, and Fig. 3 shows a thin rust-like defect that appears over a relatively wide area. 4 is a waveform diagram showing an inspection signal due to a thin rust-like defect, and FIG. 5 is a schematic diagram showing an example of an incidental part.
Figure 6 is a side view of the roll, Figure 7 is a detailed view of the optical system components, Figure 8 is a plan view of the fixed slit, Figure 9 is a plan view of the rotating slit, and Figure 10 is the electrical system component. Figure 11 is a block diagram that can be replaced with a part of Figure 10, Figure 12 is a diagram showing the contents of one cycle from loading to discharge of the object to be inspected, and Figure 13 is a diagram showing the contents of one cycle from loading to discharge of the object to be inspected. Figure 14 is a diagram showing the positional relationship between the thin line image and the rotating slit, Figure 15 is a waveform diagram showing the output signal of the photomultiplier tube, and Figure 16 is a diagram showing the preliminary measurement signal and the integration circuit. FIG. 17 is a diagram showing the relationship between the output during preliminary measurement and the inspection signal, and the relationship between the inspection signal and the defect detection level during actual measurement. 20...Photomultiplier tube, 22,22t...
・Comparator, 23...Gate circuit, 26
...Oscillator, 27 ...Counter, 2
8...Peak hold circuit, 29...
Level adjuster, 30...Divider, F, F'...
...Defect detection level.

Claims (1)

[Claims] 1. In a device that detects surface flaws on an object by photoelectrically converting light reflected from the surface of the object to be inspected and comparing the output signal with a defect detection level, the output signal from a photomultiplier tube is Preliminary measurement signals (1), (2) divided by oscillator
, (3) and a counter that sends each divided preliminary measurement signal (
Analog gate a_1 that is enabled to operate cyclically in 1), (2), and (3) and passes the output signal amplified by the divided measurement times t_1, t_2, and t_3.
, a_2, a_3, and the analog gate a
Output signals B_1, B_2, B_ of _1, a_2, a_3
3, and the integral value is used as the preliminary measurement signal (2), (3)
, (1), and the preliminary measurement signals (3), (1), (
Integrating circuits b_1, b_2, b_3 reset in 2)
and the outputs C of the integrating circuits b_1, b_2, b_3 are enabled to operate with the preliminary measurement signals (2), (3), and (1)
Analog gate d_ that passes _1, C_2, and C_3
1, d_2, d_3, and a peak hold circuit that holds the maximum output of the past integral value while being enabled to operate at the start of preliminary measurement and being reset at the completion of measurement, and the maximum value of the peak hold circuit is set to the level. It is equipped with a comparator that sets a defect detection level with a regulator and compares the defect detection level with an output signal from a photomultiplier tube, and a gate circuit that passes the output signal of the comparator for only the actual measurement time. A device for detecting surface flaws on an object, characterized by: 2 In a device that detects surface flaws on an object by photoelectrically converting light reflected from the surface of the object to be inspected and comparing the output signal with the defect detection level, the output signal from the photomultiplier is divided by an oscillator. Measurement signal (1), (2)
, (3) and a counter that sends each divided preliminary measurement signal (
Analog gate a_1 that is enabled to operate cyclically in 1), (2), and (3) and passes the output signal amplified by the divided measurement times t_1, t_2, and t_3.
, a_2, a_3, and the analog gate a
Output signals B_1, B_2, B_ of _1, a_2, a_3
3, and the integral value is used as the preliminary measurement signal (2), (3)
, (1), and the preliminary measurement signals (3), (1), (
Integrating circuits b_1, b_2, b_3 reset in 2)
and the outputs C of the integrating circuits b_1, b_2, b_3 are enabled to operate with the preliminary measurement signals (2), (3), and (1)
Analog gate d_1 that passes through _1, C_2, and C_3
, d_2, d_3, and a peak hold circuit that holds the maximum output of the past integrated value while being enabled to operate at the start of preliminary measurement and being reset at the completion of the measurement, to control the output signal from the photomultiplier tube. A divider divides the value by the maximum value of the peak hold circuit in order to align it with the reference level, and then a comparator compares it with the defect detection level set at a constant level, and the output signal of the comparator is divided by the maximum value of the peak hold circuit. A device for detecting surface flaws on an object, characterized by comprising a gate circuit that only allows passage through the surface of an object.