JPS6223804B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a method for checking whether the internal pressure of a sealed container, such as a canned food, with a metal lid, which is stored in a case that is successively conveyed by a conveyor or the like, is maintained within a predetermined range. The present invention relates to a method for automatically and continuously repeatedly testing the internal pressure of a sealed container.

Conventionally, in the internal pressure inspection of sealed containers, some automatic mechanical inspection methods have been used instead of manual inspection methods. A force is applied in a pulse manner to vibrate the lid, and the resulting damped natural vibration of the lid is detected by acoustic or electromagnetic means, converted into an electrical signal, and then processed and determined.

By the way, when a large number of sealed containers such as canned goods are arranged in a case, such as a cardboard box, and the internal pressure of the sealed containers is to be inspected from the outside while they are stored in the same case, all of the sealed containers must be inspected from the outside. Regarding the position, it is necessary to accurately detect and inspect the center point of each sealed container lid. However, the positions of the sealed containers within the case are not necessarily arranged in precise positions, but rather are considered to have positional errors in most cases. If the above-mentioned vibration is applied and the vibration is detected on the axis line at a point that is off the center position, the detected waveform will be different from that when the inspection is performed on the vertical line at the exact center position, and the detection judgment will be different. The results may also show different values, which is a factor in deteriorating test performance.

With regard to such positional deviation, an example of the results of frequency spectrum analysis of the detected waveform during canning for a sample of a gas-free canned beverage is illustrated. Figure 1 shows a case where the internal pressure shows normal negative pressure. , In the same figure, a is a waveform detected at the top of a position 10 mm off from the center of the lid, b is a waveform detected at the top of a position 5 mm away from the center of the lid, and c is a waveform detected at the top of the center of the lid. .

Next, Fig. 2 shows the case of a defective can in which the internal pressure of the can is zero, that is, the pressure is the same as atmospheric pressure.
Similarly, for c, the positional deviation is 10 mm, 5 mm, and
This is the waveform in case of 0.

Comparing both Figures 1 and 2, it can be seen that the frequency of the fundamental frequency f ₁ indicating the maximum level is constant regardless of positional deviation, and both are used as inspection data for the internal pressure of cans. The level of other large peak values, such as _f2 , changes greatly with the positional shift. Although not shown in the figure, in some cases f ₂ may exhibit a higher level than f ₁ , and in a state of positional deviation, a method of determining the internal pressure by the maximum peak frequency is used as a means of detecting the internal pressure, or With the conventional method of comprehensively measuring within a wide band of frequencies, there is a possibility that a canister without negative pressure or a canister lacking negative pressure may be mistakenly determined to be a positive pressure canister.

Next, FIG. 3 shows a case where the internal pressure of the can, which should normally be a negative pressure, is in a positive pressure, that is, an expanded state, for some reason. The absolute value of the pressure is exactly the first value.
This is a sample example that is the same as the can in the figure, and the fundamental frequency f ₁ shows the same frequency as in Figure 1, so
There is a risk that the pressure will be determined as normal, and there is a possibility that the can may be overlooked even though it should definitely be detected as a defective can.

Therefore, as a characteristic of the expansion can, we focused on the existence of the fourth harmonic of the peak waveform, as shown in Figure 3c, and we focused on the presence of a harmonic that is about 2 to 3 times the fundamental frequency, as shown in Figure 3c.
For example, it has been found that detecting a frequency _f4 around 2.4 to 2.6 times can be used as a means of discrimination. However, when inspecting with positional deviation, f ₄
The level of change changes, is almost inversely proportional to the size of the misalignment, and may even disappear in some cases.
As mentioned above, it is uncertain whether the canned food is inflated or under normal negative pressure due to the positional shift of the canned food that may occur within the case. Therefore, when inspecting sealed containers inside a case, it is clear that detection that eliminates positional deviation is necessary, but it is not easy to accurately detect the positions of many can lids arranged inside a case, and it is not practical. The method is to magnetically detect the bodies of the cans arranged at both ends of the case and assume the position of the cans inside, or to detect the outside surface of the case to calculate the position of the cans. There are indirect detection methods that determine the location of each can, but they are not always accurate.

This invention has been made with the above points in mind, and in order to prevent false detection due to positional deviation from the center position of the lid of the sealed container, the sealed container is indexed by a method such as detecting the outer surface of the case. The detection operation is performed at several positions near the assumed center position of the lid, and from among the detection signals, the value closest to the center position signal is selected to determine the quality of the internal pressure of the sealed container. be.

Next, this invention will be explained in detail with reference to the drawings from FIG. 4 showing one embodiment thereof.

In this embodiment, a case is shown in which the can-shot detection operation on the lid of a sealed container is performed three times.

First, it will be explained with reference to FIG. 4 that the detection position error can be minimized by performing the detection operation several times on the same sealed container, that is, the can to be inspected. Note that the arrow in the figure indicates the traveling direction of the can to be inspected.

In the figure, 1 is the outer periphery of the lid of the can to be inspected, 2 is the center point of the lid, 2,2'' is the regular repeated inspection point that should be inspected with pitch x, and 3 is the point caused by positional deviation. This is a positional deviation inspection point, and if the number of inspections n is 3 and the pitch x of the repeated inspection position is 5mm, the maximum allowable positional deviation to keep the detected position error within x/2 = 2.5mm is ± 1 = x/2 x n = ± 7.5 m/m. Therefore, if the positional deviation of the can in the direction of movement of the case is kept within 15 m/m, the position error will be within 2.5 m/m. It turns out.

It is possible to keep the positional deviation to this extent, and if the determination process is devised, the inspection position error can be further increased, and the allowable positional deviation can also be increased. In addition, to prevent the case from shifting in the horizontal direction relative to the direction of travel, by installing guides on both sides of the case and applying some compression pressure, the gaps between the cans can be minimized and the positional error can be kept within practical positional errors. It's easy.

Next, an example of a method for obtaining valid data from detected waveforms that have been repeated several times will be described.

In FIG. 5, the case 4 containing the canned goods to be inspected is sent over the slide plate 5 by the angle 7 connected to the drive chain 6, and when it reaches the lower part of the detection head TS, the position detection light source L The photoelectric relay blocks the projected beam of
An encoder for pulse transmission that operates PHRy and is connected to the sprocket shaft 8 of the drive chain 6.
The transmission pulse of PE is pulsed by the timing control circuit TC. With this pulse signal,
A pulse current is passed from the can-pulse current generator TP to the can-shape coil of the detection head TS, and the same detection head TS
At the same time, the detection head TS detects the attenuated waveform of the vibration of the can lid to obtain a can-shot detection electric signal. This detection signal is amplified by a preamplifier PA, and then passes through a low frequency band pass filter L/BPF, and a detector _D1 detects the frequency _f1 and level PL of the fundamental frequency component from the detection signal.

A reference frequency band f _s to _{f s} ′, which has been experimentally determined in advance according to the type of canned food to be detected, is set by a setter S ₁ , and compared with the detection fundamental frequency f ₁ by a comparison circuit C ₁ , f _s f ₁ f _s ', that is, if it is within the set frequency range, it is determined that the can is good; if it is not within the range, a defective signal is sent to the determination circuit G as an insufficient internal pressure.

Further, the detection signal of the can lid amplified by the preamplifier PA is selected by the arithmetic circuit E from among the signals obtained through the high-frequency bandpass filter H/BPF, and the fundamental frequency f ₁ and the constant α (2.0~
Find the harmonic frequency f ₄ = αf ₁ obtained by multiplying the specified number (a specific number of 3.0), and from this f ₄ and the above-mentioned high frequency component, detect a high frequency signal corresponding to f ₄ by detector D ₂ . Extract the level and obtain the harmonic signal level PH. This level PH and the level of the fundamental frequency component
A calculation unit F calculates the ratio PH/PL to PL, and a comparison circuit _C2 compares the levels of the ratio γ preset by the ratio setter _S2 and the calculation result PH/PL, As a result, if γ<PH/PL,
An excessive harmonic signal is transmitted to the determination circuit G as an expansion can.

The above is a case where the lid of a can is inspected once.
Since the same can lid is inspected twice, the detection head
With respect to the central axis of the TS, the pulse output from the pulse encoder PE is applied to the can opening pulse current generator TP and the gate by opening the gate three times by the timing control circuit TC, at approximately the center of the lid of the can to be inspected. The determination circuit G is caused to transmit a test signal, performs determination processing from among the three detection signals, selects the signal determined to be closest to the center position, and performs a pass/fail determination according to the determination conditions. The contents of this judgment process and judgment conditions are determined by the type of can and the inspection mechanism, and an example thereof will be explained here with reference to FIG. 6.

In detecting the fundamental frequency component f ₁ , two large peak levels close to the lowest frequency of the output signal of the low frequency bandpass filter L/BPF are detected.
PLm and PLm′ are detected by detectors Dm and Dm′, and the peak level PLm near the lowest frequency is set by the setter Sps.
is compared with the reference level Ps set by the comparator Cp, and if it exceeds the reference level Ps, the peak level PLm
The frequency FLm of is detected by the detector Df ₁ and set as the fundamental frequency f ₁ , and the frequency f ₁ and the level PL are detected by the comparison circuit C ₁ , the arithmetic circuit E and the arithmetic unit F in FIG.
At the same time, the ratio PLm'/PLm of the peak level PLm' of the higher frequency FLm' to this peak level PLm is determined by the arithmetic unit _FPL .

Then, the one with the smallest ratio is selected from among the three detection signals and used as the data closest to the center position of the lid of the can to be inspected. The above-mentioned FIGS. 1 to 3 show this relationship,
In both cases, c is the minimum or zero for PLm'/PLm.

Note that when the ratio PLm'/PLm is zero, for example, as in the case shown in Figure 1c, only the maximum peak level signal is detected, and the harmonic of the second number next to this is detected by the detector Dm'. This is the case if the

In addition, in the three detections, if all of the comparisons by the comparator Cp are below the reference level Ps (PS
PLm), the inspected can is considered defective (reverse can, removed can).

In the detection signal obtained in this way at the most central position, for example, in a waveform that includes a large number of harmonics f ₄ such as the example of the expansion can shown in Fig. 3, the level of the harmonic f ₄ is the highest, and This shows that Fig. 3c can be detected more reliably than Fig. 3a, which has a large deviation.

In this way, the signal closest to the center of the can lid is selected, and from the determination result of this signal in FIG. 5, when the fundamental frequency is not within the set frequency range,
Or, even if the fundamental frequency is within the set frequency range, if the ratio of harmonic components is greater than the set value,
It is possible to issue a rejection signal as a defective can and display a defective can.

As described above, we have shown an example in which it is possible to detect the internal pressure and expansion of a sealed container even if there is some positional shift, but for example, in the case of a metal container such as a can,
Significant deformation of the lid or body, or deformation due to poor seaming may also generate a large amount of harmonics, and this method may be effective as a method for inspecting defective cans.

The discrimination example described above can also be used to display or record data on inspected sealed containers, or to select the most reliable data from repeatedly inspected data as inspection data when making statistics on a large number of data. can be used.

The above data is based on a magnetic detection method for detecting vibrations of the metal lid, but similar waveforms can be obtained even when using a microphone by using an appropriate filter for external noise.

As described above, the method for inspecting the internal pressure of a sealed container according to the present invention is applicable to a sealed container having a metal lid and housed in a case that is transported by a conveyor or the like.
Magnetic pulses are applied from outside the case to several different positions on the lid of the same sealed container to vibrate the lid, and the natural damped vibrations of the lid are detected using a microphone or electromagnetic detector. , extract each low frequency component and harmonic component from each detected signal, detect two large signal levels in the low frequency component of each signal, set the lower frequency as the fundamental frequency, and Comparing and determining the ratio of both signal levels and selecting the signal closest to the center position in the lid,
Comparing the fundamental frequency of the signal with a set reference frequency range to determine whether the internal pressure is normal or insufficient,
The signal level of the harmonic component of the signal is compared with a set reference level to determine whether the internal pressure is normal or inflated.

Therefore, according to the present invention, for a sealed container in a case that is conveyed by a conveyor or the like, the can detection signal of each of several positions on the lid is used to detect the position closest to the center of the lid. Not only can the detection position be selected, but also the internal pressure of the sealed container can be determined based on this detection signal, and the negative effects of position detection errors with respect to the center position of the sealed container can be minimized, ensuring good detection accuracy. It is something that can be obtained.

[Brief explanation of the drawing]

Figures 1 to 3 are frequency spectrum analysis results of the vibration detection signal of the canned lid during canning, with each figure a being 10 m/m, each b being 5 m/m,
Each c is a waveform when the position is shifted by 0 m/m, and each signal level is shown in decibel ratio. Figure 4 and the following drawings show one embodiment of the method for inspecting the internal pressure of a sealed container according to the present invention, Figure 4 is a plan view of the can to be inspected, arrows indicate the direction of travel of the can, Figure 5 is a block diagram, FIG. 6 shows the determination conditions. 1... Lid outer periphery, 2... Center position, 2', 2''...
...Regular repeated inspection points to be inspected with pitch x, 3... Positional deviation inspection points caused by positional deviation.

Claims (1)

[Claims] 1. Magnetic pulses are applied from outside the case to several different positions on the lid of the same sealed container, which is housed in a case transported by a conveyor, etc. and has a metal lid. While vibrating the lid, the unique damped vibration of the lid is detected using a microphone or an electromagnetic detector.
The respective low frequency components and harmonic components are extracted from each detected signal, two large signal levels in the low frequency components of each signal are detected, the lower frequency is set as the fundamental frequency, and both of the above-mentioned harmonic components of each signal are detected. Comparing and determining the signal level ratio to select the signal closest to the center position of the lid, and comparing the fundamental frequency of the signal with a set reference frequency range to determine whether the internal pressure is normal or insufficient. . A method for inspecting the internal pressure of a sealed container, characterized in that the signal level of the harmonic component of the signal is compared with a set reference level to determine whether the internal pressure is normal or expanded.