JPH07133527A - Measurement of yarn - Google Patents

Abstract

PURPOSE:To provide a process for the measurement of a yarn regardless of the traveling speed and time of a yarn by analyzing a pattern consisting of the principal wavelength of the yarn. CONSTITUTION:The weight fluctuation of a traveling yarn is detected by a yarn signal coupled by AC coupling. A zero level showing the average thickness of the yarn is set based on the yarn signal. The wave form corresponding to the period from the upward passage of the yarn signal through the zero level or thereabout to the downward passage of the signal through the zero level or thereabout is detected as a major denier unevenness of a half period. The wave form corresponding to the period from the end of the half period of the major yarn denier unevenness to the upward passage of the yarn signal through the zero level or thereabout is detected as a minor denier unevenness of a half period. The wave form composed of the major unevenness half period and the minor unevenness half period is taken as one cycle to count the pattern of the yarn signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring, in a spinning process, for example, a yarn fed from a spinning machine and being wound by a winder, or a running yarn in another spinning process.

[0002]

2. Description of the Related Art It has been conventionally known that spun yarn contains various defects. For this reason, a defect of the yarn being sent out from the spinning machine and being wound up by the winder or the yarn being running in other spinning process is detected, and the defective portion is cut and removed. In general, a running yarn is passed between a pair of electrodes that are arranged to face each other, and at that time, a yarn signal is formed based on the detected electric capacity, and a peak portion appearing in the waveform of the yarn signal is allowed. By comparing with the value, thick unevenness and /
Or the thinness is judged.

Conventionally, when detecting the yarn length, for example, the length is calculated from the winding rotation amount of the winder, and at that time, the winding diameter of the yarn, which increases in diameter as it is wound, is taken into consideration. To correct the value. Alternatively, the yarn length is calculated from the traveling speed and traveling time of the yarn, and the numerical value is corrected in consideration of the change in traveling speed.

[0004]

DISCLOSURE OF THE INVENTION The present invention has found that when fluctuations in the weight of a running yarn are detected as a yarn signal that is AC-coupled, it is possible to recognize a repetitive pattern. However, as a result of knowing that it is advantageous to measure the yarn length and detect the yarn defect based on such a pattern, it is possible to provide a novel yarn measuring method that has never been seen before. Is an issue.

[0005]

A change in the weight of a yarn during traveling movement is electrically detected to form an AC-coupled yarn signal, and zero indicating the average thickness of the yarn with respect to the yarn signal. Level ±
When 0% is set, the level above the zero level ± 0% can be considered to be the “thickness unevenness” area, and conversely, the level below it can be considered to be the “thin unevenness” area.

The waveform of the AC-coupled yarn signal alternates between both thick and thin regions such as a sine wave across the zero level ± 0%, and has a waveform with an irregular amplitude. Become. Such a yarn signal is recognized as a signal in which a small wavelength component is superimposed on a certain fundamental wavelength, and the fundamental wavelength is, for example, 6 to 8 cm in cotton yarn (generally, the wavelength of the signal is frequency f It is expressed as a time component with T (sec) = 1 / f for (Hz), but it should be noted that the wavelength here is independent of time).

In other words, the amplitude and wavelength of the actual yarn signal are irregular, but the waveform thereof reaches a peak value (mountain) exceeding the zero level ± 0% and then descends to zero. Crosses level ± 0%, further descends, and the opposite peak value (valley)
It can be recognized as a continuous basic pattern in which the level rises from the zero level ± 0% toward the next mountain again after reaching 0.

Therefore, each such pattern is
By performing individual and continuous evaluations, various analyzes can be performed and yarns can be measured.

In particular, in the past, it was indispensable to convert the yarn length into time by running the yarn at a constant speed. Therefore, when the yarn is accelerated or decelerated, the measurement becomes inaccurate. While there is a drawback that the measurement itself has to be interrupted while the speed is changing or the speed is changing, the method of the present invention for individually and continuously analyzing the pattern does not include a time component. Therefore, it is particularly advantageous in that the traveling speed of the yarn can be ignored in the analysis.

(Constant Length Measuring Method) As described above, each pattern of the thread signal individually takes various values in its length (that is, wavelength), but its average value is the fundamental wavelength of the thread signal. Matches Therefore, if this pattern is counted, the length of the run yarn can be measured from the product of the counted value and the fundamental wavelength. In this case, since the pattern does not include a time component, the constant length measuring method according to this method is not affected by the change in the running speed of the yarn as seen in the conventional constant length method. Be advantageous.

In view of the above, the present invention, which is mainly intended to measure the fixed length of the yarn, is configured as the first means. The fluctuation of the weight of the yarn during traveling is detected as a yarn signal which is AC-coupled, and A zero level that indicates the average thickness of the thread is set for the thread signal, and the waveform from passing upward near the zero level of the thread signal to passing downward near the zero level is defined as the thick unevenness half cycle. In addition to detecting, the waveform from the end of the thick nonuniformity half cycle to the upward passing near zero level is detected as the thin nonuniformity half cycle, and the waveform including the thick nonuniformity half cycle and the thin nonuniformity half cycle is defined as one cycle. The point is to count the signal patterns.

Further, the present invention, which is mainly intended to measure the fixed length of the yarn, is configured as the second means, and the fluctuation of the weight of the yarn during traveling is detected as an AC coupled yarn signal, A thick unevenness level + S% and a thin unevenness level -S% are set across the zero level indicating the average thickness of the yarn signal, and the thick unevenness level + S% is set after passing the zero level upward in the yarn signal. While detecting the waveform from the rising point passing through the zero level to the falling point passing the zero level downward after passing the zero level downward as a thick unevenness half cycle,
The waveform from the falling point to the rising point passing through the thick unevenness level + S% after passing the zero level upward again is detected as a thin unevenness half cycle, and a waveform composed of the thick unevenness half cycle and the thin uneven half cycle is 1 The point is that the pattern of the thread signal is counted as a cycle.

(Method of Detecting Moire) Moire is known as one of the problems that unevenness of yarn affects the appearance of the woven fabric. The fundamental wavelength component of the yarn is originally stronger than any other superimposed wavelength component, but when it becomes relatively stronger, this fundamental wavelength becomes more noticeable in the fabric, and as a result, the fabric looks like a wood grain pattern. Will give a nice look. This is called moire. Such a fabric naturally has no commercial value.

When a yarn which causes such moire is examined by a yarn unevenness tester (trade name: Evenestesta), it is confirmed that the peak and thin peak values of the pattern are larger than those of the normal yarn. It

That is, moire is a state in which patterns of large amplitude are continuous. Therefore, an appropriate defect level of ± a% is set in the thick unevenness area and the thin unevenness area, and when both the thick and thin peak values of the pattern exceed the ± a%,
If the sum is continuously calculated as 1 or -1 otherwise, it is considered that the sum rapidly becomes a large positive value when moire appears (however, the sum is negative). When it becomes, it is desirable to fix it to 0).

Therefore, if a positive value is set in advance and some kind of alarm is issued when the sum reaches this value, it is possible to detect a yarn defect that causes moire. Even in this case, there is an advantage that the traveling speed of the yarn is not affected when the moire is detected.

Therefore, according to the present invention, which is mainly configured to detect moire, the weight fluctuation of the yarn during traveling is detected as a yarn signal coupled in an alternating current, and the average of the yarn is detected with respect to the yarn signal. While setting zero unevenness level + S% and thin unevenness level -S%,
Thick side defect level + a% exceeding the unevenness level + S%
And the fine unevenness level-fine side defect level below S%-
a% is set; from the ascending point of passing the thick level + S% after passing the zero level upward of the thread signal to the descending point of passing the thin uneven level −S% after passing the zero level downward. The waveform is detected as a thick unevenness half cycle, and the waveform from the falling point to the rising point passing through the thick unevenness level + S% after passing upward through the zero level again is detected as a thin unevenness half cycle, and the thick unevenness half cycle is detected. The pattern of the yarn signal is counted with one cycle consisting of a waveform consisting of a cycle and a fine nonuniformity half cycle; when a pattern exceeding both the thick side defect level + a% and the thin side defect level -a% is detected, the count is counted up, The point is to count down when a pattern that does not exceed at least one of the side defect level + a% and the fine side defect level -a% is detected.

(Continuous slab detection method) One of the causes for lowering the evaluation of the woven fabric is the continuous slab of yarn. This is a series of three or four slab defects that appear at short intervals. Even if each slab is not so big, it is conspicuous in the fabric because it is continuous.

In detecting such a continuous slab, first, + b is used to determine the thickness of the slab to be detected.
Set a slab level of% in the thick spot area. Then, when the thick peak value of the pattern exceeds the slab level + b%, it is set to +1.

By the way, if the slab level + b% is not exceeded, the slab level + b% is not immediately set to -1.
It is set to -1 when n patterns which do not exceed% are consecutive. Here, the meaning of n is to give a maximum interval for considering two slabs as continuous, and the interval is set by the number of patterns. In this way, if the sum is continuously calculated, the sum may rapidly become a large positive value when a continuous slab appears (however, when the sum becomes negative, Should be fixed to 0). Even in this case, there is an advantage that the traveling speed of the yarn is not affected when the continuous slab is detected.

In view of the above, the present invention is mainly configured to detect continuous slabs. The fluctuation of the weight of the yarn during traveling is detected as an AC-coupled yarn signal, and the yarn average is calculated with respect to the yarn signal. The uneven thickness level + S% and the thin unevenness level −S% are set across the zero level indicating the thickness, and the slab level + b exceeding the uneven thickness level + S% is set.
%; Waveform from the rising point of passing the thick level + S% after passing the zero level upward of the thread signal to the falling point of passing the thin uneven level −S% after passing the zero level downward. Is detected as a thick unevenness half cycle, and a waveform from the falling point to the rising point passing through the thick unevenness level + S% after passing through the zero level upward again is detected as a thin unevenness half cycle, and the thick unevenness half cycle is detected. And the pattern of the thread signal is counted with one cycle of a waveform consisting of a half cycle of fine unevenness; when a pattern exceeding the slab level + b% is detected, the pattern is counted up, and n patterns which do not exceed the slab level + b% are continuously obtained. The point is to count down when it is detected.

[0022]

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

(Pattern Recognition and Constant Length Measuring Method) The above-mentioned thread signal pattern is preferably thick with a point crossing upward with respect to the zero level ± 0% indicating the average thickness of the thread as a starting point. After passing through the unevenness area, the waveform up to the point where zero level ± 0% is crossed downward toward the thin unevenness area is defined as a thick unevenness half cycle, and the thick unevenness is obtained after passing through the thin unevenness area from the end point of the thick unevenness half cycle. Zero level ± 0% toward the area
Assuming that the waveform up to the point intersecting upward is a thin nonuniformity half cycle, a pattern in which a waveform composed of such a thick nonuniformity half cycle and a thin nonuniformity half cycle is one cycle is recognized.

However, the actual thread signal is usually several mV.
Since there is a white noise of several tens of mV, a certain amount of hysteresis is necessary to judge the crossing of the signal zero level ± 0%. Therefore, this level is set to ± S%.

FIG. 2 illustrates the pattern of the thread signal. In the illustrated example, a region thicker than + S% is a thick unevenness region, and a region thinner than -S% is a thin unevenness region. + S
The region between% and -S% is the dead zone. Therefore, + S%
Is called a thick unevenness level, and -S% is called a thin unevenness level.

The points (the starting point and the ending point of the thick nonuniformity half cycle and the starting point and the ending point of the thin nonuniformity half cycle) which are the delimiters for the pattern recognition are in the dead zone from the thin nonuniformity area to the thick nonuniformity area or in the opposite direction. It's time to go in through. If this is not the case, for example, when the signal in the thick nonuniformity region once falls to the dead zone and then enters the thick nonuniformity region again, it is not used as a breakpoint for recognizing the pattern.

That is, as shown in FIG. 2, the thread signal 1 passes through the thick unevenness region from an ascending point 2 which passes through the thick unevenness level + S% after passing through the zero level ± 0% upward, and The waveform up to the falling point 3 that passes through the thin unevenness level −S% after passing the zero level ± 0% downward is recognized as a thick unevenness half cycle. As described above, the superimposed waveform 4 which enters the dead zone from the thick area but returns to the thick area again is ignored from the recognition of the pattern.

Further, the waveform from the descending point 3 to the ascending point 2 which passes through the narrow unevenness region and upwardly passes the zero level ± 0% and then passes the thick unevenness level + S% is also a thin unevenness half. Recognize as a cycle. Although not shown, in this case as well, the superimposed waveform that enters the dead zone from the thin uneven area but returns to the thin uneven area again is ignored from the pattern recognition.

Therefore, one pattern is recognized with the basic waveform consisting of the thick nonuniformity half cycle and the thin nonuniformity half cycle as one cycle, and thereafter, the patterns continue.

FIG. 1 is a block diagram of a circuit for recognizing a pattern. A specific example for pattern recognition will be described with reference to the signal timing chart of FIG.

The binarization circuit 5 shown in FIG. 1 has a yarn signal 1 which is AC-coupled and a reference voltage + for binarization.
S% and -S%, and the sampling clock 6 are input. The output of the binarization circuit 5 is 2 corresponding to + S%.
The binarized signal (thickness unevenness binarized signal 7) and the binarized signal (thin unevenness binarized signal 8) corresponding to -S% are output in synchronization with the sampling clock 6.

The binarization binarization signal 7 becomes H when the yarn signal 1 exceeds the bins level + S% and is in the bins area, and the flip-flop 9 is set at the rising edge thereof. When the flip-flop 9 is set, the output (thickness unevenness half-cycle signal 10) becomes H.

On the other hand, the thin unevenness binarized signal 8 becomes H when the signal 1 exceeds the thin unevenness level -S% and is in the thin unevenness area, and the flip-flop 11 is set at the rising edge thereof. When the flip-flop 11 is set, the output (fine unevenness half-cycle signal 12) becomes H, and the flip-flop 9 is reset at the rising edge thereof. Therefore, the thickness unevenness half-cycle signal 10 becomes L, and the thickness unevenness half-cycle signal 10 ends the recognition of the thickness unevenness half-cycle of the pattern in the thread signal 1.

Subsequently, the thread signal 1 passes through the narrow unevenness area, then enters the dead zone again, passes upward through the zero level ± 0%, and crosses the thick unevenness level + S% to enter the thick unevenness area. Then, the flip-flop 9 is set again, and the thick nonuniformity half-cycle signal 10 becomes H. At that time, the thick nonuniformity half-cycle signal 10 resets the flip-flop 11 at the rising edge, so the thin nonuniformity half-cycle signal 12 becomes L.
Thus, the thin nonuniformity half cycle signal 12 ends the recognition of the thin nonuniformity half cycle of the pattern in the thread signal 1.

The thick unevenness half-period signal 10 and the thin unevenness half-period signal 12 set the flip-flops 13 and 14 at their respective rising edges. But this 2
The two flip-flops 13 and 14 are constantly reset in synchronization with the falling edge of the sampling clock 6. Therefore, the thick unevenness start timing signal 15 is output from the flip-flop 13, and the thin unevenness start timing signal 16 is output from the flip-flop 14, and the clock of the thick unevenness start timing signal 15 indicates the start point of the thick unevenness half cycle of the thread signal 1. The clock of the thin unevenness start timing signal 16 indicates the starting point of the thin unevenness half cycle of the thread signal 1. In other words, this makes it possible to detect the rising point 2 and the falling point 3 of the yarn signal 1.

As described above, for example, the basic wavelength (pattern length) of the cotton yarn is 6 to 8 cm, and the basic wavelength of the yarn has a certain degree of regularity depending on the fiber length and the like, so the thick unevenness start timing signal 15 and / or By counting the thin nonuniformity start timing signal 16, it becomes possible to measure the length of the thread that has been run, regardless of time.

(Moire Detection Method) FIG. 3 is a block diagram of a circuit for detecting moire, and a specific example for moire detection will be described with reference to the signal timing chart of FIG.

In the binarization circuit 21 shown in FIG. 3, the yarn signal 1 which is AC-coupled and the reference voltage sensitivity ± a for detecting moire are set.
%, And the sampling clock 22 is input.

The thin unevenness half-period signal 12, the thick unevenness half-period signal 10, the thick unevenness start timing signal 15, and the thin unevenness start timing signal 16 are output from the circuit for pattern recognition described with reference to FIGS. 1 and 2. Since the above-mentioned circuit may be applied, each signal is denoted by the same reference numeral, and the description of the circuit is omitted.

The thick-side defect signal 23 indicates that the peak value of the thick nonuniformity half cycle exceeds the moire thick sensitivity + a% in one pattern. On the other hand, the narrow-side defect signal 24 is, on the contrary, 1
It is shown that the peak value of the fine nonuniformity half cycle of the two patterns is lower than the Moire fine sensitivity −a%. The thick-side defect signal 23 and the thin-side defect signal 24 are respectively supplied to the binarization circuit 2.
The flip-flops 25 and 26 are set by the output from 1 and reset by the thin nonuniformity half-cycle signal 12 and the thick nonuniformity half-cycle signal 10, respectively.

The half-cycle defect pattern signal 27 is flip-flop 28 set by the thick defect signal 23.
And is cleared at the start point of the next pattern by the thick unevenness start timing signal 15. Therefore, the half-cycle defect pattern signal 27 is a signal for storing that the peak value of the thick unevenness half cycle in one pattern exceeds the moire thick sensitivity + a%.

The full-cycle defect pattern signal 29 is stored in the memory circuit 30 set by the narrow-side defect signal 24. The half-cycle for storing that the peak value of the thick unevenness half cycle in a certain pattern exceeds the moire thick sensitivity + a% is stored. Using the defect pattern signal 27 as data, the narrow side defect signal 24 output when the peak value of the fine nonuniformity half cycle in the pattern exceeds the Moire fine sensitivity −a% is output as the set trigger signal, and the next pattern of the next pattern is output. It is cleared by the thin unevenness start timing signal 16 indicating the start point of the second half period (fine unevenness half period). Therefore, the full cycle defect pattern signal 29
Is a signal indicating that the peak value exceeds the moire sensitivity ± a% in both the first half period (thickness unevenness half period) and the latter half period (thin unevenness half period) of the pattern.

The full cycle defect pattern signal 29 is sent to the U / D control terminal (up / down) of the up / down counter 31 at the next stage.
The down-thickness start timing signal 15 is input to the count clock terminal of the counter 31.

Therefore, when the peak values of the first half cycle (thickness unevenness half cycle) and the second half cycle (fine unevenness half cycle) of one pattern both exceed the moire sensitivity ± a%, the full cycle defect pattern signal 29 is generated. Since it becomes H, the up / down counter 31 performs up-counting (additional counting) according to the thick unevenness start timing signal 15 input to the counting clock.

By the way, the yarn signal exceeds the moire sensitivity + a% in the first half cycle of the pattern (thickness unevenness half cycle), and even if the half cycle defect pattern signal 27 is H, the moire sensitivity in the latter half cycle (fine unevenness half cycle). If -a% is not exceeded,
Since the memory circuit 30 is not cleared by the thin unevenness start timing signal 16 and is not triggered by the narrow side defect signal 24, the full cycle defect pattern signal 29 is L.
On the contrary, the first half cycle of the pattern (half cycle of thick unevenness)
In the case, the yarn signal does not exceed the moire sensitivity + a%, but in the latter half cycle (fine unevenness half cycle), the moire sensitivity exceeds -a%.
Since the half-cycle defect pattern signal 27 in the L state is stored, the full-cycle defect pattern signal 29 is L as well.

Therefore, the full cycle defect pattern signal 29 is L
In this case, the up / down counter 31 performs down-counting (subtraction counting) by the thick unevenness start timing signal 15 input to the counting clock. When the up / down control terminal is L and the count value inside the counter is zero, the zero detection 34 is output. This terminal is connected to the reset terminal and sets the counter internal numerical value to zero in priority to the subtraction counting. Therefore, when at least one of the peak values of the first half cycle (thickness unevenness half cycle) and the second half cycle (fine unevenness half cycle) of one pattern does not exceed the moire sensitivity ± a%, the internal value of the counter is fixed to zero.

As a result of the above, the counted 4-bit output indicating the numerical value of the up / down counter 31 is compared with the value set by the operator in the next numerical value comparison circuit 32, and if they match, the moire alarm signal 33 is output. To be done.

(Continuous Slab Detection Method) FIG. 5 is a block diagram showing a circuit for detecting a continuous slab. A concrete example for continuous slab detection will be described with reference to the signal timing chart of FIG. .

In the binarization circuit 41 shown in FIG. 5, the yarn signal 1 which is AC-coupled and the reference voltage sensitivity + b for slab detection are added.
%, And the sampling clock 42 is input.

The thick nonuniformity start timing signal 15 and the thin nonuniformity start timing signal 16 are output from the circuit for pattern recognition described with reference to FIGS. 1 and 2, and the above circuit may be applied. Each signal is denoted by the same reference numeral, and the description of the circuit is omitted.

When the thread signal 1 exceeds the slab sensitivity + b%,
The binarization circuit 41 outputs a binarized signal in synchronization with the sampling clock 42, and sets the flip-flop 43. The flip-flop 43 that has been set has a thick unevenness start timing signal 15 indicating the start point of the next pattern.
Is reset by. As a result, flip-flop 4
From 3 onwards, as shown in FIG. 6, the thread signal 1 is the slab sensitivity +
A slab signal 44 indicating that b% has been exceeded is output.

The slab signal 44 is supplied to the down counter 4
5 is directly input to the asynchronous preset terminal 5 and the U / D control terminal (up / down control terminal) of the up / down counter 46, and further serves as a gate signal for the count clock of the up / down counter 46.

The down counter 45 has a slab sensitivity of + b%.
This is for measuring the maximum interval between slabs exceeding the number of patterns by the number of patterns, and is set by the operator with the number of patterns as a preset value. This preset value determines the permissible range of the continuous dense and dense state of the slab for a predetermined length of yarn, and in the illustrated embodiment, this preset value (the number of patterns) n is 6. The thin nonuniformity start timing signal 16 is input to the count clock of the down counter 45. The zero detection signal 47 is input to the sync preset terminal.

When the thread signal 1 exceeds the slab sensitivity + b%,
The asynchronous preset terminal of the down counter 45 is triggered by the slab signal 44, and the internal numerical value of the down counter 45 is set to a preset value (6 in the illustrated embodiment) asynchronously with the counting clock.

On the other hand, when the thread signal 1 does not exceed the slab sensitivity + b%, the down counter 45 is set to 1 by the thin unevenness start timing signal 16 appearing after the slab signal 44 is cleared by the thick unevenness start timing signal 15.
Count down one by one. When the internal numerical value of the down counter 45 becomes zero, the zero detection signal 47 is output, and the zero detection signal 47 is cleared when the internal numerical value of the down counter 45 is set to the preset value. The setting of the internal numerical value from zero to the preset value is performed in synchronization with the counting clock, or asynchronously by the slab signal 44 as described above. The zero detection signal 47 is O
It is inputted as a counting clock 48 of the up / down counter 46 through the R gate, and is input to the up / down counter 4
Count down six.

The up / down counter 46 is for counting the number of slabs exceeding the slab sensitivity + b%. As the counting clock 48, a signal obtained by ANDing the thin unevenness start timing signal 16 and the slab signal 44 is used. , A signal obtained by taking the logical sum of the zero detection signal 47 is input. Since the slab signal 44 is directly input to the U / D control terminal (up / down control terminal) of the up / down counter 46, the up / down counter 46 detects that the thread signal 1 exceeds the slab sensitivity + b%. While the addition 1 is performed for each pattern, if the slab sensitivity + b% is not exceeded, the subtraction 1 is performed when the number of patterns that do not exceed the preset value of the down counter 45 (6 in the illustrated embodiment) continues.

In this way, the counted 4-bit output indicating the numerical value of the up / down counter 46 is compared with the value set by the operator in the next numerical value comparison circuit 49,
If they match, the continuous slab alarm signal 50 is output.

The up / down counter 46 has a zero detection output and is input to the reset terminal. Therefore, if a slab exceeding the slab sensitivity + b% is not detected over a long yarn section, the up / down counter 46
Is reset in preference to the subtraction count and is fixed at zero.

[0059]

According to the first aspect of the present invention, by counting the pattern of the yarn signal, the yarn is analyzed regardless of the traveling speed and time of the yarn, and the yarn length is measured and
There is an effect that it is possible to detect a yarn defect such as a moire or a continuous slab.

According to the second aspect of the present invention, when counting the pattern of the thread signal, for example, the signal having the thick unevenness region is set to a basic wavelength such that the signal once falls to the dead zone and then returns to the thick unevenness region. There is an effect that small wavelength components that have been superimposed can be eliminated and only the pattern of the fundamental wavelength can be counted.

According to the third aspect of the present invention, there is an effect that the moire can be detected from the pattern of the yarn signal regardless of the traveling speed of the yarn.

According to the present invention described in claim 4, there is an effect that a continuous slab can be detected from the pattern of the yarn signal regardless of the traveling speed of the yarn.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a circuit for recognizing a pattern according to the present invention.

FIG. 2 is a timing chart of signals based on the circuit of FIG.

FIG. 3 is a block diagram showing an embodiment of a circuit for detecting moire in the present invention.

FIG. 4 is a timing chart of signals based on the circuit of FIG.

FIG. 5 is a block diagram showing an embodiment of a circuit for detecting a continuous slab according to the present invention.

6 is a timing chart of signals based on the circuit of FIG.

[Explanation of symbols]

 1 Thread signal 7 Thick uneven binarization signal 8 Thin uneven binarization signal 10 Thick uneven half period signal 12 Thin uneven half period signal 15 Thick uneven start timing signal 16 Fine uneven start timing signal 23 Thick side defect signal 24 Narrow side defect signal 27 Half-cycle defect pattern signal 29 Full-cycle defect pattern signal 31 Up / down counter 44 Slab signal 45 Down counter 46 Up / down counter

Claims (4)

[Claims] 1. A change in the weight of a yarn during traveling is detected as a yarn signal coupled in an alternating current, and a zero level indicating the average thickness of the yarn is set for the yarn signal, and zero of the yarn signals is set. The waveform from passing upward in the vicinity of the level until passing downward in the vicinity of the zero level is detected as a thick uneven half cycle, and from the end of the thick uneven half cycle to passing upward in the vicinity of the zero level. Is detected as a thin nonuniformity half cycle, and the pattern of the thread signal is counted with the waveform consisting of the thick nonuniformity half cycle and the thin nonuniformity half cycle as one cycle. 2. A weight variation of a yarn during traveling movement is detected as a yarn signal coupled in an alternating current, and a thickness unevenness level + S% is sandwiched by a zero level indicating an average thickness of the yarn with respect to the yarn signal. Fine thinness level -S% is set, and after passing the zero level of the thread signal upward, it passes thick unevenness level + S%. After passing zero level downward from the ascending point, it passes thin fineness level -S%. The waveform up to the falling point is detected as a thick unevenness half cycle, and the waveform from the falling point to the rising point passing through the thick unevenness level + S% after passing through the zero level again again is detected as a thin uneven half cycle. A yarn measuring method, characterized in that a pattern of a yarn signal is counted with a waveform consisting of the thick unevenness half cycle and the thin unevenness half cycle as one cycle. 3. A weight variation of a yarn during traveling movement is detected as a yarn signal which is AC-coupled, and a thickness unevenness level + S% is sandwiched by a zero level indicating an average thickness of the yarn with respect to the yarn signal. The thin unevenness level −S% is set, and the thick side defect level + a% exceeding the thick unevenness level + S% and the thin side defect level −a% lower than the thin unevenness level −S% are set.
Half-cycle of the yarn signal from the rising point passing through the thick unevenness level + S% after passing the zero level upwards to the falling point passing through the thin unevenness level -S% after passing the zero level downwards And the waveform from the falling point to the rising point passing through the thick unevenness level + S% after passing through the zero level upward again, is detected as the thin unevenness half cycle, and the thick unevenness half cycle and the thin uneven half cycle are detected. The pattern of the yarn signal is counted with the waveform consisting of 1 cycle as one cycle; when a pattern exceeding both the thick side defect level + a% and the thin side defect level -a% is detected, the count is up-counted, and the thick side defect level + a% and The main purpose is to detect moiré, which is characterized by counting down when a pattern that does not exceed at least one of the fine side defect level -a% is detected. Method of measuring the thread that. 4. A weight variation of a yarn during traveling movement is detected as a yarn signal coupled in an alternating current, and a thickness unevenness level + S% is sandwiched by a zero level indicating an average thickness of the yarn with respect to the yarn signal. The thin unevenness level −S% is set, and the slab level + b% exceeding the thick unevenness level + S% is set; from the rising point at which the thin unevenness level + S% is passed after passing through the zero level of the thread signal upward. The waveform up to the falling point passing through the thin unevenness level −S% after passing the zero level downward is detected as a thick unevenness half cycle, and the unevenness level + S% is passed after passing the zero level upward again from the falling point. The waveform up to the rising point passing through is detected as a thin uneven half cycle, and the waveform of the thick uneven half cycle and the thin uneven half cycle is taken as one cycle to count the pattern of the thread signal; the slab level Measurement of a yarn mainly for continuous slab detection, which is characterized by counting up when a pattern exceeding + b% is detected and counting down when detecting n patterns which do not exceed the slab level + b% continuously. Method.