JP6022249B2 - Rotary weight sorter - Google Patents

Description

  The present invention relates to a rotary weight sorting apparatus.

  As this kind of weight sorting device, a plurality of weighing devices are provided around the rotating body, and the weight sorting device that measures the weight of the object to be weighed supplied to the weighing device while rotating the rotating body and sorts it by weight rank. (For example, refer to Patent Document 1).

  FIG. 17 is a plan view of the weight selection device of Patent Document 1. In this rotary type weight sorting apparatus, a plurality of, for example, 16 loads S1 to S16 and loads for detecting the loads S1 to S16 are detected around a turntable 60 that rotates around a rotation center O. A measuring instrument comprising a sensor 61 is provided, and is rotated in the direction of arrow A (counterclockwise) at a predetermined rotational speed together with the turntable 60.

  The objects to be weighed 13 are placed on the supply conveyor 4 that moves straight in the left direction indicated by the arrow B in synchronization with the rotation speed of the platforms S1 to S16, and is partitioned by bars provided at predetermined intervals. Placed in the area.

  The movement of one crosspiece is configured to be synchronized with the rotation movement of 1/16 of one of the platforms S1 to S16 rotating on the circumference around the rotation center O, and is mounted on one mounting area. The objects to be weighed 13 are sequentially supplied onto one of the platforms S1 to S16.

  In FIG. 17, the platforms S <b> 1 to S <b> 16 to which the object to be weighed 13 is supplied near the lower end a of the conveying end of the supply conveyor 4 rotate in the direction of arrow A with the object to be weighed 13 being placed. An arithmetic circuit (not shown) in the control box 62 on the turntable 60 converges the transient response signal of the load signal from the load sensor 61 that detects the weight of the platforms S1 to S16 on which the object 13 is placed. Waiting for stabilization, when the platforms S1 to S16 reach the vicinity b immediately before the sorting position (1) to which the weighing object 13 is first sorted, a stable load signal is read and the weight of the weighing object 13 is read. Calculate the value. Based on the calculated weight value and the preset boundary weight value, the weight rank of the object 13 is determined. Then, according to the determined weight rank, the objects to be weighed 13 are discharged and distributed to the corresponding distribution positions (1) to (8) in the discharge range shown in FIG.

  Now, for example, the platform S16 is weighed at a position b immediately before the first distribution position (1) shown in FIG. 17, and the weight rank is determined by the arithmetic circuit in the control box 62. When the weight rank (7) is assumed, the stage S16 rotates while the weight rank data is stored in the arithmetic circuit, and the stage S16 reaches the center of the distribution position (7) in FIG. Thus, the double-open gate of the mounting table S16 is completely opened, and the object 13 is distributed to the distribution position (7).

JP-A-5-31464

  In Patent Document 1, the weights of the objects to be weighed 13 on all the platforms S1 to S16 are respectively measured in the vicinity b immediately before the first distribution position (1), and the weight ranks are respectively determined. Is measured in the vicinity b just before the weighing object of the weight ranks (2) to (8) corresponding to the distribution positions (2) to (8) behind the first distribution position (1). The weight rank thus determined is held, and the object to be weighed 13 is simply conveyed to the corresponding distribution positions (2) to (8).

  In general, as long as the drift of the zero point or the like is small, it is well known that the weight can be accurately measured as a long time elapses after the object to be weighed is supplied to the platform.

  On the rotation circumference of the platforms S1 to S16, the weighing object 13 is measured at least once before the weighing object 13 reaches the first distribution position (1) from the supply position of the weighing object 13, and the weighing object is measured. Although it is necessary to determine whether or not the object 13 has the weight rank (1) corresponding to the first distribution position (1), the weight rank corresponding to the distribution positions (2) to (8) behind the object 13 is determined. It is not appropriate in terms of weighing accuracy to determine the weight rank of the weighing object 13 only from the first measured weight value. To increase the weighing accuracy, the measurement position for measuring the weight from the supply position of the weighing object 13 Should be as long as possible.

  The present invention has been made in view of the above points, and an object of the present invention is to improve weighing accuracy in a rotary weight sorting apparatus.

  In order to achieve the above object, the present invention is configured as follows.

(1) The rotary weight sorting apparatus according to the present invention includes a driving unit that rotates a measuring instrument along a circumference around a center of rotation, and the weighing in which an object to be weighed is supplied at a supply position on the circumference. Determination means for measuring the weight of the object to be weighed based on a load signal from a container and determining a weight rank to which the object to be weighed belongs among a plurality of weight ranks. a plurality of distribution locations corresponding individually are arranged along the circumference, of the distribution positions of the multiple, the distribution position corresponding to the weight rank is determined by the determination means, the object to be weighed A rotary weight sorting device for sorting,
Said plurality of distributing positions, along said direction of rotation of the measuring instrument to be weighed is fed by the supply position, the supply position as the upstream, are arranged toward the downstream side of the supply position, The determination means includes the load signal when the weighing instrument to which the object to be weighed is supplied has reached a first measurement position upstream of the plurality of sorting positions arranged along the circumference. The weight rank is determined by measuring the weight of the object to be weighed, and further, the weighing machine to which the object to be weighed is distributed at the plurality of distribution positions until the object to be weighed is distributed, When one or more measurement positions are reached, the weight rank is determined by measuring the weight of the object to be weighed based on the load signal at that time .

The first measurement position, Ru upstream der than sorting position on the most upstream side.

Before SL one or more measurement positions can be a single measurement position, but preferably a plurality of measurement positions so as to correspond to a plurality of weight classes. In addition, the time from supplying the object to be weighed to measuring its weight is made as long as possible so that the transient response of the load signal is smaller and converges and stabilizes. It is preferable that the measurement position is further on the downstream side so as to ensure time for smoothing vibration noise due to disturbance.

According to the rotary weight sorting apparatus of the present invention, when the weighing instrument supplied with the object to be measured at the supply position reaches the first measurement position, the weight of the object to be weighed is measured based on the load signal. rank is determined, further, until objects to be weighed is distributed, when it reaches the one or more measurement positions in a plurality of distributing position on the downstream side of the first measurement position, As a measurement position, i.e., the first Compared to the measurement position, a long time has passed since the supply of the object to be weighed, and the transient response of the load signal converges to a smaller and stable weight. Compared to the conventional example in which the weight of the object to be weighed is measured only once in a short time and the weight rank is determined, the weighing accuracy of the entire object to be weighed can be improved.

(2) In a preferred embodiment of the rotary weight sorting apparatus according to the present invention, the sorting position corresponding to the weight rank determined by the weight of the weighing object measured at the first measurement position is provided with a plurality of the weighing instruments. , toward the downstream side of the meter, until distributes the objects to be weighed, the greater the number of reachable said one or more measurement positions.

According to this embodiment, the distribution position corresponding to the weight rank determined by the weight of the measurement object measured at the first measurement position reaches one or more that reaches the distribution object in the downstream scale. since the number is a number of measurement positions, it is possible to determine the weight classes and weighed more measurement positions on the downstream side. At the measurement position on the downstream side, the transient response of the load signal converges and stabilizes after a long time has elapsed from the supply of the object to be weighed, and the weight value is obtained using the long time load signal. Thus, the smoothness with respect to the disturbance vibration noise signal generated during the measurement period can be increased, so that the weighing object having the weight rank corresponding to the distribution position on the downstream side can be measured with higher accuracy.

(3) In the embodiment of the above (2), the first measurement position is just before the weighing instrument supplied with the object to be measured reaches the most upstream first distribution position among the plurality of distribution positions. a position, before Symbol one or more measurement positions, as well as corresponding to the distribution position of the downstream side of the first sorting position, a position immediately before reaching the vibrating dividing position, the determining means, each distribution The weight rank may be determined by measuring the weight of the object to be weighed based on the load signal from the weighing instrument every time the measurement position immediately before the position is reached.

  The position immediately before reaching the distribution position is when the weight rank is determined by measuring the weight immediately before this position, and the weight rank is the weight rank corresponding to the distribution position that is reached immediately after In addition, it refers to a position where an object to be measured can be distributed to the distribution position. That is, even if the weight rank of the weighing object whose weight has been measured is the weight rank of the sorting position that arrives immediately after that, it is a position where the weighing object can be assigned to the sorting position.

  According to this embodiment, the weight is measured and the weight rank is determined each time the weighing instrument supplied with the weighing object reaches each measurement position immediately before each sorting position. The weight rank is determined based on the weight measured immediately before the distribution position corresponding to the rank, and is distributed to the distribution position that is reached immediately thereafter. In other words, since the weight is measured at the measurement position immediately before the distribution position where the object is actually distributed, the time when the transient response of the load signal due to the supply of the object to be measured converges and stabilizes, and the weight The load signal used for obtaining the value can be secured to the maximum, and the weighing accuracy can be further increased.

 (4) In the embodiment of the above (3), when the weight rank determined based on the measured weight of the weighing object is a weight rank corresponding to the distribution position that the measuring instrument has already passed. Based on the measured weight of the object to be weighed, the weight rank may be determined to be one lighter weight or one heavier than the determined weight rank.

  Multiple weight ranks are classified according to the boundary weight value, but in the case of an object to be weighed having a weight near the boundary weight value, the light side classified according to the boundary weight value due to the influence of vibration noise, etc. Or, it may be determined as one of the weight ranks of both weight ranks on the heavy side. For example, a weight rank that is one lighter than the true weight rank or a weight rank that is one heavier than the true weight rank is determined. The vessel may pass through.

  According to this embodiment, when the determined weight rank is a weight rank corresponding to the distribution position that the weighing instrument has already passed, one weight rank or one lighter than the determined weight rank based on the measured weight. It can be re-determined to a heavy weight rank.

 (5) In the embodiment of the above (4), the objects to be weighed in the measuring device may be distributed to the distribution position corresponding to the weight rank re-determined by the determining means.

  According to this embodiment, the objects to be weighed can be distributed to the distribution positions corresponding to the re-determined weight rank.

  In this case, since it can be re-determined and assigned to one light weight rank or one heavy weight rank, the measured weight value of the object to be weighed is the upper limit value of one light weight rank or the lower limit value of one heavy weight rank. It is preferable that the weight value of the measured object is matched with the weight rank to be distributed.

  According to the present invention, not only when the weighing instrument reaches the first measurement position, but also after a long time has elapsed from the supply of the object to be weighed, the transient response of the load signal converges and stabilizes less, and the long time The weight rank is determined by measuring the weight even when it reaches the downstream measurement position where the weight value can be obtained from the load signal, so the weight rank is determined by measuring the weight of the object to be measured only once. Compared to the conventional example, the measurement accuracy can be increased.

FIG. 1 is a plan view showing a schematic configuration of a rotary weight sorting apparatus according to an embodiment of the present invention. FIG. 2 is a diagram showing a plurality of weight ranks and their boundary weight values. FIG. 3 is a block diagram showing a control configuration of the weight sorting apparatus. FIG. 4 is a diagram schematically showing the configuration of the filter. 5A and 5B are diagrams showing the configuration of the pulse generator, where FIG. 5A is a plan view and FIG. 5B is a partially longitudinal side view. 6A and 6B are diagrams showing pulses, where FIG. 6A shows a reset pulse and FIG. 6B shows a timing pulse. FIG. 7 is a diagram showing the relationship between the system state and the position of the mounting table. FIG. 8 is a diagram showing a platform on which weight is to be acquired and an output point of a filter to be acquired in each system state. FIG. 9 is a diagram showing a platform on which the necessity of driving the cylinder that opens and closes the gate should be verified in each system state. FIG. 10A is a flowchart for explaining the operation, and is a flowchart showing the highest priority processing. FIG. 10B is a flowchart showing processing subsequent to FIG. 10A. FIG. 11 is a flowchart showing details of step n8 in FIG. 10A. FIG. 12 is a flowchart showing a process for obtaining a weight value between the central control unit and each measurement unit. FIG. 13 is a flowchart showing an outline of processing according to the system state. FIG. 14A is a flowchart showing the processing of the system state P1. FIG. 14B is a flowchart showing processing subsequent to FIG. 14A. FIG. 14C is a flowchart showing processing subsequent to FIG. 14B. FIG. 14D is a flowchart illustrating processing subsequent to FIG. 14C. FIG. 14E is a flowchart showing processing subsequent to FIG. 14D. FIG. 14F is a flowchart showing processing subsequent to FIG. 14E. FIG. 15 is a diagram showing another example of the filter format. FIG. 16 is a diagram schematically showing the configuration of a filter according to another embodiment of the present invention. FIG. 17 is a schematic plan view of a conventional example.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(Embodiment 1)
FIG. 1 is a plan view showing a schematic configuration of a rotary weight sorting apparatus according to an embodiment of the present invention.

The weight sorting apparatus 1 of this embodiment includes a turntable 2 that rotates around a rotation center O in the direction of arrow A at a predetermined rotation speed, and rotates around the turntable 2 together with the turntable 2. A plurality of, in this example, 16 mounts S1 to S16 are provided, and on the turntable 2, a control box 8 containing a centralized control unit and the like to be described later is provided. Each loading table S1 to S16 is supported by a lower portion of the turntable 2 by the load sensor 5 ₁ to 5 ₁₆ consisting of a load cell (LC) for detecting the load of the loading table S1 to S16, each load platform S1~ the S16 and the load sensors 5 ₁ to 5 _16, constitute sixteen meter.

  An object to be weighed 13 such as a fishery product or a marine product is dropped and supplied from the supply conveyor 4 onto the platforms S1 to S16. In the supply conveyor 4, the placement areas on which the objects 13 are individually placed are partitioned at predetermined intervals by bars or the like, and the objects 13 are conveyed in the direction indicated by the arrow B. One of the rotating platforms S1 to S16 and one of the placement areas of the supply conveyor 4 operate in synchronization with each other, and one object 13 placed on each placement area is supplied. Sequentially supplied to the platforms S1 to S16 that have been rotated to the supply position below the conveyance end of the conveyor 4.

  In the article supply range shown in FIG. 1, the weight of the object 13 supplied from the supply conveyor 4 to the platforms S <b> 1 to S <b> 16 is measured as described below while rotating, and the measured weight value and a preset value are set. Based on the boundary weight value, the weight rank is determined.

  In this embodiment, as shown in FIG. 2, based on the measured weight value Wx and the boundary weight values Wc1 to Wc7, the weight rank of the object 13 is set to eight weights as follows. It is determined as one of ranks (1) to (8).

・ Wx <Wc1 Weight rank (1)
・ Wc1 ≦ Wx <Wc2 Weight rank (2)
・ Wc2 ≦ Wx <Wc3 Weight rank (3)
……………………………………………………………
・ Weight rank (7) when Wc6 ≦ Wx <Wc7
・ When Wc7 ≦ Wx Weight rank (8)
The to-be-measured object 13 from which the weight rank was determined is placed on the platforms S1 to S8 at the distribution positions (1) to (8) corresponding to the determined weight rank in the discharge range along the circumferential direction shown in FIG. The gate of S16 is opened and discharged and distributed.

  In addition, in the following description, after the to-be-measured object 13 is supplied to the mounting bases S1-S16 in a supply position, it is from distribution to the distribution position (1)-(8) corresponding to the weight rank of the to-be-measured object 13. In the rotation path of the platforms S1 to S16, the upstream side in the rotational direction (side closer to the supply position) is referred to as the front, and the downstream side in the rotational direction (side away from the supply position) is referred to as the rear.

Conventionally, as described above, on the rotational circumference of the platforms S1 to S16, each platform S1 to S1 is reached by the time it reaches the distribution position (1) provided first from the supply position of the object to be weighed. S16 based on the load signal from the load sensor 5 ₁ to 5 ₁₆ to measure the weight of the objects to be weighed to determine the weight rank, so that distributes respectively distributing position corresponding to the determined weight classes. That is, before reaching the distribution position (1), the weight is measured once, based on the measured weight value, one of the weight ranks (1) to (8) is determined, and the weight rank is determined. After the determination, the weight rank is maintained, and the object to be weighed is distributed at the corresponding distribution position.

On the other hand, in this embodiment, in order to improve the weighing accuracy, each of the platforms S1 to S16 arrives in front of the distribution positions (1) to (7) of each rank. S16 was measured each weight by the load signal from the load sensor 5 ₁ to 5 ₁₆ to determine the weight rank, and the objects to be weighed 13 to distribute respectively distributing position of the determined weight classes. When reaching the position before the distribution position (8), it is not necessary to measure the weight. This means that if it is determined that the weight rank (6) or the weight rank (7) is not the weight rank (7) as will be described later by the weight value when it reaches before the distribution position (7), the remaining weight is not measured. This is because the weight rank (8), which is the weight rank of

Thus as each loading table S1~S16 reaches the front of the distribution position of each weight categories (1) to (7), the load signal from the load sensor 5 ₁ to 5 ₁₆ of each loading table S1~S16 Based on the above, the weight is measured to determine the weight rank, and the objects to be weighed 13 are assigned at the assigned weight rank assignment positions. Therefore, the assignment is performed on the rear side among the assignment positions (1) to (7). The measured weight rank of the object to be weighed is measured at the measurement position on the rear side, and the weight rank is determined. Therefore, as the objects to be weighed weight ranks of the rear side, load from the articles 13 increases the time from being supplied to the load platform S1~S16 until measuring its weight, the load sensor 5 ₁ to 5 ₁₆ The transient response of the signal will be smaller and converge and stabilize, and the weight value can be obtained using a long time load signal, that is, the load signal that has passed through a filter with a large smoothing characteristic as described later Since the weight value can be obtained from the above, the measurement accuracy as a whole can be improved.

  For example, if the weight rank of the weighing object 13 supplied to the mounting table S1 is the weight rank (1) corresponding to the distribution position (1), the mounting table S1 has reached before the distribution position (1). Sometimes, the weight is measured to determine the weight rank, the weight rank (1) is determined, and the distribution is made at the distribution position (1).

  However, for example, if the weight rank of the weighing object 13 supplied to the mounting table S1 is the weight rank (4) corresponding to the distribution position (4) behind the distribution position (1), the mounting table S1. When reaching before the distribution position (1), when reaching before the distribution position (2), when reaching before the distribution position (3), and when reaching before the distribution position (4) The weight is measured to determine the weight rank, and the weight is determined to be the weight rank (4) based on the weight measured before reaching the distribution position (4), and the distribution is performed at the distribution position (4).

  For example, when the weight rank of the object 13 supplied to the mounting table S1 is the weight rank (7) corresponding to the rear distribution position (7), the mounting table S1 is at the distribution position (1). When reaching the front, before reaching the sorting position (2), when reaching before the sorting position (3), when reaching before the sorting position (4), before the sorting position (5) When reaching, before reaching the distribution position (6), and when reaching before the distribution position (7), the weight is measured to determine the weight rank, and before the distribution position (7). Is determined as a weight rank (7) based on the weight measured when it reaches, and is distributed at the distribution position (7).

  In this way, as the weighing object 13 having the weight rank at the rear distribution position has a longer time from the supply of the weighing object 13 to the weight measurement, the transient response signal converges to be smaller and the stable weight. The value can be obtained.

Further, the weight values that are obtained by measuring in front of each sorting location, converts the analog signal from the load sensor 5 _{1-5 16} described above into a digital signal, although the reading from the load signal smoothed by the filter In this embodiment, smoothing processing is performed using a filter having a response characteristic corresponding to the time required for each of the platforms S1 to S16 to reach each distribution position from the supply position of the object to be weighed.

  That is, on the rotation circumference of the platforms S1 to S16, the rearward distribution position away from the supply position of the object to be weighed has more time until the weight measurement, so smooth processing is performed with a filter having a large response characteristic. I am doing so.

  Thus, the weight rank is determined by obtaining a stable weight value from a load signal that has passed through a filter having a large smoothing characteristic, that is, a load signal with a long response time, for an object to be weighed at the rear distribution position. Can do.

  In this embodiment, as shown in FIG. 1, in order to determine the weight rank by acquiring the weight value of the object to be weighed 13 before each distribution position and determine the weight rank in this embodiment, as shown in FIG. .. 8 are virtually determined along the rotation circumference of the platforms S1 to S16, which are sections that divide the semicircular circumference into eight equal parts. Since the mounting bases S1 to S16 are arranged along the rotation circumference, each of the outer peripheral positions 1, 2, 3,... 8 is a section that can correspond to one mounting base.

  The outer peripheral position 1 is located one section ahead of the first distribution position (1), and the outer peripheral positions 2 to 8 are positions corresponding to the distribution positions (1) to (7), respectively.

  In FIG. 1, boundaries between the outer peripheral positions 1 to 8 are indicated by virtual extension lines d1 to d8, and intermediate positions at the outer peripheral positions 1 to 7 are indicated by virtual extension lines I1 to I7.

  FIG. 3 is a block diagram showing a control configuration of the weight sorting apparatus 1 of FIG.

Analog load signal of the load sensor 5 ₁ to 5 ₁₆ for detecting a load of the load platform S1~S16 above respectively, are input to the respective measurement units ₆₁ through _16.

Each of the measurement units 6 _{1 to} 6 ₁₆ amplifies the analog load signal output from the load sensors 5 _{1 to} 5 ₁₆ and also includes an amplifier 7 incorporating an analog filter for removing high frequency noise, an arithmetic circuit 9, And a serial controller 10.

  The arithmetic circuit 9 includes an A / D converter (not shown) that converts an analog load signal output from the amplifier 7 into a digital load signal, and the A / D converted load signal is weighted as described later. A CPU and a memory (not shown) that perform filtering according to the measurement position are provided.

16 units load signals generated by the respective measurement unit ₆₁ through ₁₆ of the serial centralized control unit 11 from the serial controller 10 provided in each of the measuring units ₆₁ through ₁₆ via a serial line SL1 It is sent to the controller 12.

The centralized control unit 11 includes a calculation control circuit 18 including a CPU and a memory (not shown). The calculation control circuit 18 includes timings from first and second photosensors PH1 and PH2, which will be described later, constituting a pulse generator. A pulse Tp and a reset pulse Rp are given via the I / O circuit 16, and the rotational positions of the platforms S1 to S16 are recognized based on these pulses Tp and Rp. The central control unit 11 collects the load signals of all the platforms S1 to S16 from the measurement units 6 _{1 to} 6 ₁₆ and measures the weight at a predetermined timing, that is, obtains the weight value.

  In addition, the arithmetic control circuit 18 of the central control unit 11 acquires the weight value, determines the weight rank, and discharges and distributes the object to be measured 13 at the distribution position corresponding to the weight rank. The cylinders cy1 to cy16 that open and close the gates of the mounting platforms S1 to S16 are controlled via the. Further, the arithmetic control circuit 18 of the centralized control unit 11 controls a motor 19 as a driving means via the I / O circuit 16 and the motor drive circuit 17, and thereby the rotation table 2 driven to rotate by the motor 19. The rotation speed is controlled to the set rotation speed.

  The arithmetic control circuit 18 of the central control unit 11 transmits the acquired weight value, the determined weight rank, and the like to the display setting unit 21 outside the rotating body via the serial controller 20 of the central control unit 11 to display the display setting unit. In 21, the serial controller 22 receives the data via the serial line SL2.

  Communication between the central control unit 11 on the rotation side and the display setting unit 21 on the fixed side is performed by a rotary connector (not shown), and the power supplied to the display setting unit 21 side is for power supply (not shown). Power is supplied to the central control unit 11 side through the slip ring. The central control unit 11 and the display setting unit 21 may be provided with a wireless transmission / reception circuit for wireless communication.

  The display setting unit 21 includes an arithmetic control circuit 23 including a CPU and a memory (not shown), an input unit 24 provided with various switches for setting a boundary weight value, a rotation speed, and the like for weight rank determination, A display unit 25 for displaying data and the like sent from the control unit 11 and an I / O circuit 26 are provided.

In this embodiment, as described above, each measurement is performed so that the weight value can be acquired from a filter having a large smoothing characteristic as the time from the supply of the object 13 to the platforms S1 to S16 to the measurement of the weight value is longer. Units 6 _{1 to} 6 ₁₆ are configured.

Next, a description will be given of each measuring unit ₆₁ through _16.

  The weight sorting apparatus 1 can be operated for weighing by setting various rotational speeds. For example, the rotation speed at the minimum weighing capacity is 10 rpm, and the 16 platforms S1 to S16 rotate per rotation, so that the object to be weighed can be placed on all the platforms S1 to S16. The weighing processing capacity is a rotation speed corresponding to 160 pieces / min.

  Further, assuming that the rotation speed at the maximum weighing capacity is 20 rpm and the objects to be weighed can be placed on all the platforms S1 to S16, the weighing processing capacity is a rotation speed corresponding to 320 pieces / minute. .

  The rotation speed is set directly as the rotation speed per minute or indirectly as the value of the weighing capacity per minute.

  Here, the position on the main body from which the weight value is acquired will be described.

  In FIG. 1, the position on the circumference for obtaining the weight value for determining the weight rank of the weighing object 13 supplied from the supply conveyor 4 is the distribution position of the weighing object 13 in this embodiment ( The positions of the above-described virtual extension lines I1, I2, I3,..., I7 slightly before 1) to (7) are set. The virtual extension lines I1, I2, I3,... I7 are arranged on the circumference at intervals of 1/16 of one round section of the rotation circumference of the platforms S1 to S16.

  However, the distribution position (8) is a position that is distributed when the weight value of the object to be weighed acquired at the position of the virtual extension line I7 does not belong to the weight rank (7) as a result of the weight rank determination.

  Further, in FIG. 1, a position that is four sections ahead of the virtual extension line I1, which is a weight value acquisition position necessary to distribute the object 13 at the first distribution position (1) (one section is the circumference) (1/16 length section) is defined as a virtual extension line I0. The virtual extension line I0 is a position in the vicinity where the object 13 is supplied.

  From virtual extension lines I0 to I1, from virtual extension lines I0 to I2,..., Each distance from virtual extension lines I0 to I7, that is, each distance given by a ratio to the circumference of one rotation of the platform, This is referred to as a weight measurement interval when the virtual extension lines I1, I2,.

In this embodiment, each measuring unit ₆₁ through _16, but the fixed filter and a variable filter for filtering the load signal converted A / D is provided, it will be described first fixed filter.

FIG. 4 schematically shows a shift register and a block configuration of arithmetic processing provided in each arithmetic circuit 9 of each of the measurement units 6 _{1 to} 6 ₁₆ .

Here, by the case where the objects to be weighed 13 when the rated capacity of the objects to be weighed 13 is not placed on the platform S1~S16 is placed, the platform S1~S16 and load sensors 51 _to It is assumed that the natural frequency of the weighing instrument having 5 ₁₅ changes between 30 and 33 Hz.

The arithmetic circuit 9 of each of the measurement units 6 _{1 to} 6 ₁₆ corresponding to each of the platforms S1 to S16 includes an A / D conversion circuit 27 that converts the analog load signal output from the amplifier 7 into a digital load signal. I have. The A / D conversion circuit 27 converts the analog load signal into a digital load signal at a sampling interval of 1 msec, for example.

  The arithmetic circuit 9 is provided with a first shift register SFR1 composed of 30 serial cell registers. Each cell register stores one A / D sampling data that has been A / D converted by the A / D conversion circuit 27.

  Each time one A / D sampling data which is a measured value of the load signal is newly obtained from the A / D conversion circuit 27 every 1 msec, the data of the cell register is sequentially shifted to the right in FIG. By discarding one A / D sampling data at the right end and inputting the latest A / D sampling data, the latest 30 A / D sampling data are always stored.

  Further, every time one new A / D sampling data is stored in the first shift register SFR1, the average value of the latest 30 cell register data stored in the first shift register SFR1 It is calculated by the average value arithmetic circuit (AV1) 30 and given as input data to the second shift register SFR2.

  The second shift register SFR2 includes 33 serial cell registers, and is connected to the next stage of the first average value arithmetic circuit (AV1) 30.

  Each time the average value is newly calculated by the first average value arithmetic circuit (AV1) 30 at an interval of 1 msec, the second shift register SFR2 sequentially shifts the data in the cell register one by one in the right direction in FIG. The latest average value for 33 is always stored and the oldest average value at the right end is discarded.

  Each time a new average value is stored in the second shift register SFR2 at an interval of 1 msec, the second average value calculation circuit (AV2) 31 in the next stage is connected to the latest 33 of the second shift register SFR2. The average value of the cell register outputs is calculated.

  The first and second average value calculation circuits (AV1, AV2) 30, 31 and the first and second shift registers SFR1, SFR2 are not specially provided, and the average calculation data in the calculation circuit 9 is not provided. This is a schematic representation of the flow and calculation, and is executed in the CPU and the memory connected to the CPU.

  The first shift register SFR1 composed of 30 cell registers and the first average value arithmetic circuit (AV1) 30 serve as a notch filter for a noise signal having 30 msec as one cycle, and the second shift register SFR2 composed of 33 cell registers. The second average value calculation circuit (AV2) 31 becomes a notch filter for a noise signal having one cycle of 33 msec, and becomes a double notch filter.

The double notch filter is a filter for the natural vibration noise removal meter comprising a load platform S1~S16 and load sensor 5 ₁ to 5 _15, when a load signal includes a noise signal 30～33Hz, namely When one period includes vibration noise of 33 msec to 30 msec, a large damping effect is obtained.

  The supply position of the object to be weighed 13 on the circumference varies depending on the placement state of the object to be weighed 13 on the supply conveyor 4 and the difference in adhesiveness.

  In this embodiment, in the case of the rotational speed at the maximum weighing capacity, the platforms S1 to S16 are supplied from the platforms S1 to S16 after the object 13 is supplied on the platforms S1 to S16, and the virtual extension line I0 in FIG. The time required to reach the position is set to a response time of the double notch filter of 63 msec or more. That is, the position of the virtual extension line I0 corresponds to the positions of the platforms S1 to 63 after the elapse of 63 msec or more after the object to be weighed 13 is supplied on the platforms S1 to S16 latest in the case of the rotational speed at the maximum weighing capacity. The position reached by S16.

  As another embodiment of the present invention, the position at which the object to be weighed is supplied on the platforms S1 to S16 the latest is the position of the virtual extension line I0, and thereafter the object to be weighed reaches the virtual extension line I1. The response time of the natural vibration noise removing filter may be expected to be 63 msec within the elapsed time up to.

Since this double notch filter, it is an object of the smoothing of the natural vibration noise meter with a load platform S1~S16 and load sensor 5 ₁ to 5 _15, rotation speed and weight of the platform S1~S16 It is necessary to provide regardless of the acquisition position of the value.

Note that, the gate of the platform S1 to S16, the gate mounting bracket, if so, tare weight applied to the load sensor 5 ₁ to 5 _16, such as a cylinder for gating is sufficiently large relative to the weight of the objects to be weighed are Since the change in the natural vibration period of the measuring device due to the presence or absence of the objects to be weighed on the platforms S1 to S16 is small, only one notch filter of the double notch filters may be provided.

  Next, in addition to the above-described notch filter that is a fixed filter, a variable filter that changes the smoothing characteristic according to the rotational speed of the turntable 2 and the weight acquisition position will be described.

  The platforms S1 to S16 are moved in accordance with the rotational movement from the virtual extension line I0 to I1, I2,... I7 in FIG. 1, virtual extension lines I01 to I1, virtual extension lines I0 to I2, virtual extension lines I0 to I3, ... the virtual extension lines I0 to I7 and the weight measurement interval gradually increase, and the required time gradually increases at the same rotational speed.

  In this embodiment, a filter is provided for each of the virtual extension lines I1, I2, I3,... I7, depending on the value of the rotation speed and their weight acquisition positions, that is, the respective weight acquisition positions from the position of the virtual extension line I0. A filter having a large or small response time, that is, a variable filter having a large and small smoothing characteristic, is applied according to the time required for moving the platforms S1 to S16.

  As a result, a filter having a larger smoothing characteristic is applied to the load signal at the rear weight value acquisition position on the rotation circumference of the platforms S1 to S16 so that an accurate weight value can be acquired.

  First, when the rotation speed is the highest, that is, when the maximum rotation speed is 20 rpm, the time required for one rotation is 3 seconds. Since the required time per section of 1/16 of one round is 3000/16 = 187.5 msec, in FIG. 1, the platforms S1 to S16 are 4 from the position of the virtual extension line I0 to the position of the virtual extension line I1. The required time for moving the section is 187.5 msec * 4 = 750 msec. Further, it is necessary to move through the sections of the virtual extension lines I1 to I2, the virtual extension lines I2 to I3, the virtual extension lines I3 to I4, the virtual extension lines I4 to I5, the virtual extension lines I5 to I6, and the virtual extension lines I6 to I7. The time is 187.5 msec.

  On the other hand, in the case of the minimum rotation speed, since the required time is twice, the required time for the platforms S1 to S16 to move four sections from the position of the virtual extension line I0 to the position of the virtual extension line I1 is 375 msec. * 4 = 1500 msec. Further, it is necessary to move through the sections of the virtual extension lines I1 to I2, the virtual extension lines I2 to I3, the virtual extension lines I3 to I4, the virtual extension lines I4 to I5, the virtual extension lines I5 to I6, and the virtual extension lines I6 to I7. The time is 375 msec.

  As the response time by the filter, only the required time for the object 13 to move in each section may be expected. First, in the weight measurement interval from the virtual extension lines I0 to I1, the maximum rotation speed is 750 msec, the minimum rotation speed In this case, 1500 msec can be expected. Since these times can be estimated as the maximum value of the filter response time in setting the filter, they are called allowable response times.

  The allowable response time from the virtual extension line I0 to each virtual extension line I2, I3,... I7 increases by 187.5 msec at the maximum rotation speed and by 375 msec at the minimum rotation speed.

  That is, a filter having a response time corresponding to an allowable response time that differs depending on whether the rotation speed is maximum or minimum can be cascade-connected to the fixed filter (notch filter).

  As a general expression, when the rotation speed Vrpm is given, the time required for one rotation is (60 / V) sec. Therefore, the required time per 1/16 section (= 1 section) is (60 · V). ) * (1/16) = (15/4) * (1 / V) sec = (3750 / V) msec, and the virtual extension lines I0 to I1 are 4 sections, so that the allowable response time is (15000 / V) msec can be expected.

  The virtual extension lines I1 to I2, the virtual extension lines I2 to I3,..., The virtual extension lines I6 to I7, and the required time for each section, 187.5 msec at the maximum rotation speed, 375 msec at the minimum rotation speed, In general, in the case of Vrpm, the required time is (3750 / V) msec. Therefore, the allowable response time is increased by (3750 / V) msec at the virtual extension lines I2, I3,. .

  The response time of the filter provided per section after the virtual extension line I1 can be increased in increments of (3750 / V) msec in anticipation of a longer time load signal at the rear position. Even if a vibration signal is generated, the smoothing characteristic with respect to vibration noise can be increased at the output point of the filter corresponding to the measurement position at the rear position on the circumference, and a more stable weight value can be acquired.

  As an example of the variable filter, a case where an average value filter is used will be described.

  In the second shift register SFR2 and later constituting the fixed filter of FIG. 4, the average value of the second shift register SFR2 is obtained at intervals of 1 msec, whereas many cell registers are serialized by taking the average value of intervals of 10 msec. The average value is stored in the connected third shift register SFR3.

  The method of storing data in the third shift register SFR3 is the same as that of the first and second shift registers SFR1 and SFR2. First, the data in each cell register is sequentially shifted to the right by one and then the second shift register. Store the new data obtained from SFR2. This operation is performed at 10 msec intervals.

  This is because the vibration noise removed by the variable filter has a relatively small frequency compared to the natural frequency of the measuring instrument, that is, the period is long, so a lot of memory necessary to store the load signal for a long time is used. The shift processing time can be shortened.

  As another embodiment, a large number of memories may be used to store a load signal every 1 msec.

  When the rotation speed of the weight sorting apparatus 1 is set to the minimum value of 10 rpm, the cell registers constituting the third shift register SFR3 are prepared in a number that can correspond to the minimum rotation speed. FIG. 4 shows a configuration corresponding to this minimum rotation speed of 10 rpm.

  One cell register data is sequentially shifted to the right in FIG. 4 every 10 msec, and the data is stored from the leftmost cell register having the output point I0. Therefore, in the case of the minimum rotation speed, the response time is 1500 msec. As a shift register up to the output point I1, 150 cell registers are connected.

  Furthermore, since the response time of 370 msec is expected up to the output point I2, 37 cell registers are connected. Since the same response time is expected until the output points I3, I4, I5, I6, and I7, 37 cell registers are connected in series, and a total of 150 + 37 * 6 = 372 cell registers are connected in series.

  These registers are configured in the RAM of the arithmetic circuit 9.

  On the other hand, when the maximum rotational speed is 20 rpm, the response time must be advanced, and the number of cell registers up to the output point I1 with a response time of 750 msec as described above is 75. In addition, since the number of cell registers prepared for each interval between the output points I2, I3,... I7 may be 18, a total of 75 + 18 * 6 = 183 cell registers are connected in series.

  When a rotational speed of, for example, 15 rpm is set between the rotational speeds of 10 rpm and 20 rpm, the number of cell registers connected in inverse proportion to the rotational speed is determined.

As a method of determining the number of connection registers, generally when a rotation speed, for example, 18 rpm is set,
Speed ratio: (20-18) / (20-10) = 1/5
The number of registers up to the output point I1 is 75+ (150−75) * (1/5) = 90.
The number of cell registers between the output points I1 to I2,..., And the output points I6 to I7 is 18+ (36-18) * (1/5) = 21 (the number after the decimal point is rounded down), and the total is 21 *. 6 = 126 is determined.

  As a general formula, when the value of the rotation speed V is set, the value of the rotation speed V may be replaced instead of the rotation speed 18 rpm. As the number of shift register cell registers is increased, the average value of sampling values for a long-time load signal is obtained. Therefore, as a filter, the smoothing characteristic for a long-period vibration signal is generally increased, and various shorter short-circuit values are obtained. Smoothness characteristics for periodic vibration signals are also increased.

  A filter that obtains an average value with a larger numerical value has a longer response time, but it can obtain a large smoothing characteristic for floor vibration that occurs suddenly at a weight measurement interval, and a smoothing characteristic for a transient response vibration signal is also large. .

  Next, setting of filter conditions will be described.

The rotation speed V is set from the input unit 24 of the display setting unit 21 in FIG. 3, and the measurement units 6 _{1 to} 6 ₁₆ corresponding to the platforms S _{1 to S 16} from the display setting unit 21 via the central control unit 11. Is provided via the serial line SL1.

When the rotational speed V is given, the above calculation is performed in each of the measurement units 6 _{1 to} 6 ₁₆ , and the number of cell registers corresponding to the rotational speed V and the memory address corresponding to the number are determined.

  Note that the value of the rotation speed V is not necessarily the set value itself, but the central control unit 11 measures the generation time interval between the reset pulse Rp and the timing pulse Tp from the first and second photosensors PH1 and PH2. It is good also as a rotational speed calculated based on the recognized generation time interval.

  Next, designation of the output point to the filter and output of the weight value from the filter will be described.

As shown in FIG. 1, in order to obtain the weight value of the filter in the vicinity of the positions of the virtual extension lines I1, I2,... I7 according to the rotational movement on the circumference of the platforms S1 to S16, these positions are obtained. to when the load platform S1 to S16 is reached, gives a read command of data of the filter as will be described later, the measurement unit ₆₁ through ₁₆ corresponding the central control unit 11 to a predetermined load platform S1 to S16.

  For example, assuming that the metering operation is performed at a rotational speed of 10 rpm, the code for designating the output point I1 from the central control unit 11 to the corresponding measurement unit at the timing of obtaining the weight value at the position of the virtual extension line I1. And the loading table number are given together, the measurement unit corresponding to the loading table designated as shown in FIG. 4 interprets this and interprets this for each of the 150 cell registers from output points I0 to I1. The average value of the output is calculated by the third average value calculation circuit (AV3) 32, and the set data of the data content code indicating the weight value, the number of the loading table and the average value is obtained via the serial controller 10. The weight value is sent to the central control unit 11.

  If the position of the output point I2 is designated as the weight value for reading, the average value of the outputs of 150 + 37 = 187 cell registers between the output points I0 to I2 is calculated as a third average value arithmetic circuit (AV3) 33. And is sent to the central control unit 11 as an acquired weight value.

  Similarly, the acquired weight value is output up to the output point I7.

  As described above, the weight value that has passed through the filter having a large smoothing characteristic is acquired as the output points I1, I2,... It is transmitted to the unit 11.

  Next, control in the central control unit 11 will be described.

  First, the creation of a system state for recognizing where each of the platforms S1 to S16 is located will be described.

  As shown in FIG. 1, the platforms S1 to S16 rotate on the circumference, and the output of the filter corresponding to the weight acquisition position as described above at any one of the outer peripheral positions 1 to 7 for each of the platforms S1 to S16. A weight value is acquired from the points, a weight rank is determined based on the acquired weight value, and a measurement control sequence for distributing the objects to be weighed on the platforms S1 to S16 at a distribution position corresponding to the determined weight rank is created.

  The central control unit 11 recognizes the positions of the platforms S1 to S16 on the rotation circumference based on the pulses from the pulse generator shown in FIGS. 5A and 5B, and on the circumference of the platforms S1 to S16. The system state corresponding to the position of is created in the central control unit 11.

  1 is supported by a column 28 provided at the rotation center O, and the columns 28 are also rotationally driven by the motor 19 described above, so that the platforms S1 to S16 are also rotationally driven. Is done.

  An optical sensor bar 29 is attached to the column 28, and the optical sensor bar 29 rotates around the rotation center O together with the column 28.

  A first photosensor PH1 composed of a light projecting / receiving element for generating a timing pulse Tp and a second photosensor PH2 composed of a light projecting / receiving element for generating a reset pulse Rp are attached to the tip of the optical sensor bar 29, and a fixing portion 37 is attached. Is attached with an annular ring 35. This annular ring 35 is provided with 16 timing pulse generation slits 33 at positions that divide the circumference into 16 equal parts, of which the timing pulse generation slit 33 overlaps the angle with one timing pulse generation slit 33. A reset pulse generating slit 34 that is slightly wider than the slit 33 is provided.

  When the mounting platforms S1 to S16 and the optical sensor bar 29 rotate, as shown in FIG. 6B in synchronization with the rotation of the mounting platforms S1 to S16, 16 timing pulses Tp generated per one rotation of the optical sensor bar 29; As shown in FIG. 6A, one reset pulse Rp generated per one rotation of the optical sensor bar 29 is generated and read by the central control unit 11.

  The pulse width pwr of the reset pulse Rp is wider than the pulse width pwt of the timing pulse Tp. The central control unit 11 includes a clock pulse generation circuit that generates a pulse signal whose period is sufficiently shorter than the timing pulse width pwt. The reset pulse Rp and the timing pulse Tp are read by the arithmetic control circuit 18 of the central control unit 11 by the interrupt processing program having the highest priority activated by the pulse signal from the clock pulse generation circuit.

  The circular ring 35 is installed on the fixing portion 37 of the main body so that the line a in the circular ring 35 in FIG. 5 substantially overlaps the position of the line d1 shown in FIG. The optical sensor bar 29 is provided so as to overlap the position of the end g on the rotation advance side of the mounting table S1 in FIG.

  Then, when the end g on the rotation advance side of the mounting table S1 reaches the position of the line d1, the reset pulse Rp and the timing pulse Tp1 in FIGS. 6 (a) and 6 (b) are generated. Thereafter, the end g on the rotation advance side of the mounting table S1 reaches the line d2, the mounting table S1 completely overlaps the outer peripheral position 1, and the timing pulse Tp2 is generated by the next timing pulse generating slit 33. A state until the timing pulse Tp2 is generated is defined as a system state P1.

  Further, after the generation of the timing pulse Tp2, the stage S1 is completely overlapped with the outer peripheral position 2, and the timing pulse Tp3 is generated by the next timing pulse generating slit 33. The state until this timing pulse Tp3 is generated Is defined as a system state P2.

  Similarly, every time the timing pulses Tp3, Tp4,... Tp16... Are generated, the system state P3, P4,.

  In this way, the system state proceeds every time the timing pulse Tp is generated while the platforms S1 to S16 and the optical sensor bar 29 rotate around the circumference, and proceeds to the system states P1 to P16, and again the system state P1. Return to and repeat.

  In FIG. 6, each system state is started, a time point (1) for designating an output point of a filter for obtaining a weight value of a predetermined stage, a time point (2) for reading the weight value of the designated output point, The time point (3) when the weight rank is determined and the distribution signal is held is shown together with the time point (4) when the gate opening / closing cylinder is driven by the held distribution signal.

  When the system state is started, in the central control unit 11, the first predetermined time T01 shown in FIG. 6, which is the time when the platforms S1 to S16 arrive near the positions of the virtual extension lines I1, I2,. At the time (1) when (msec) elapses, transmission specifying the output point to obtain the weight value from the predetermined filter output point of the measurement unit corresponding to the predetermined stage corresponding to the value of the system state Px Create signal data and send it to the corresponding measurement unit. Also, the weight value is transmitted from the corresponding measurement unit to the central control unit 11 and stored in the memory within the second predetermined time T0 (msec) up to the time point (2). The central control unit 11 determines the weight rank by the time point (3) based on the acquired weight value, generates and holds a distribution signal for distribution at the distribution position corresponding to the weight rank, and the next system At the start time (4) of the state, the driving of the gate opening / closing cylinder is started at the distribution position corresponding to the distribution signal.

  FIG. 7 shows which platforms S1 to S16 are located at the outer peripheral positions 1 to 8 in FIG. 1 in each of the system states P1 to P16.

  In the system state P1, the stage S1 starts to enter the outer peripheral position 1 until it completely enters and overlaps, that is, in FIG. 1, the end g on the rotation advance side of the stage S1 exceeds the virtual extension line d1. This corresponds to the period up to the extension line d2. During this period, the stage S16 is placed at the outer circumferential position 2, the stage S15 is placed at the outer circumferential position 3, the stage S14 is placed at the outer circumferential position 4, the stage S13 is placed at the outer circumferential position 5, and the stage S12 is placed at the outer circumferential position 6. The base S11 enters the outer peripheral position 7 and the mounting base S10 enters the outer peripheral position 8 and then completely enters and overlaps.

  In the system state P2, the stage S2 is at the outer peripheral position 1, the stage S1 is at the outer peripheral position 2, the stage S16 is at the outer peripheral position 3, the stage S15 is at the outer peripheral position 4, and the stage S14 is at the outer peripheral position 5. The stage S13 is placed at the outer peripheral position 6, the stage S12 is placed at the outer peripheral position 7, and the stage S11 is placed at the outer peripheral position 8, and then completely enters and overlaps.

  Similarly, as the system state advances, the platforms that enter the respective outer peripheral positions 1 to 8 are shifted one by one.

  In the state shown in FIG. 1, since the platforms S16 to S9 are completely overlapped with the outer peripheral positions 1 to 8, respectively, the system state S16 is completed as shown in FIG. Shows just before the start.

  Next, the system state and weight value acquisition timing will be described.

  In FIG. 1, in the system state P1, the platforms S1 to S16 shown in () reach the outer peripheral positions 1 to 8. That is, the stage S1 is at the outer peripheral position 1, the stage S16 is at the outer peripheral position 2, the stage S15 is at the outer peripheral position 3, the stage S14 is at the outer peripheral position 4, the stage S13 is at the outer peripheral position 5, and the stage S12. Are placed in the outer peripheral position 6, the mounting base S 11 is in the outer peripheral position 7, and the mounting base S 10 is in the outer peripheral position 8.

  For the stage S1, it is sufficient that the weight value can be obtained at the position of the virtual extension line I1 in FIG. 1. By this position, the average value of the cell register output of the shift register SFR3 up to the filter output point I1 in FIG. It only has to be read.

  For the stage S16, it is only necessary to obtain the weight value at the position of the virtual extension line I2 in FIG. 1. By this position, the average value of the cell register output of the shift register SFR3 up to the filter output point I2 in FIG. It only needs to be readable.

  Similarly, the platforms S15 to S12 need only be able to read the average values of the cell register outputs from the filter output points I3 to I6 in FIG. 4 up to the positions of the virtual extension lines I3 to I6 in FIG.

  Also for the mounting table S11, it is only necessary to obtain the weight value at the position of the virtual extension line I7 in FIG. 1, and it is only necessary to read the average value of the cell register output of the shift register SFR3 up to the filter output point I7 in FIG.

  The weight value read at the position of the virtual extension line I7 is determined as one of the weight ranks (6), (7), and (8) as described later.

  Since the weight rank (8) is determined at this point, the stage S11 no longer needs to acquire the weight value in the next system state P2, and even when the weight rank (8) is determined, the system state P1. The weight value read at the position of the virtual extension line I7 may be used as the weight value of the object to be weighed.

  Therefore, in the system state P1, it is not necessary to acquire a weight value for the mounting table S10.

  FIG. 8 shows the platforms S1 to S16 from which the weight values are to be obtained and the output points of the filters from which the weight values are to be read in the respective system states P1 to P16.

  In the system state P1, as described above, the weight values up to the filter output points I1, I2, I3, I4, I5, I6, and I7 are respectively obtained for the mounts S1, S16, S15, S14, S13, S12, and S11. Get it.

  Further, in the system state P2, if the weight values up to the filter output points I1, I2, I3, I4, I5, I6, and I7 are obtained for the platforms S2, S1, S16, S15, S14, S13, and S12, respectively. Good.

  Similarly, as the system state progresses, it is only necessary to shift one stage for obtaining the weight value.

  Based on this FIG. 8, the central control unit 11 designates one of the filter output points I1 to I7 for the measurement unit 6k corresponding to the platform Sk of the number k shown in FIG. 8 for each system state. .

Filter example, the system state P1, as shown in FIG. 8, to specify the output point I1 of the filter with respect to the measuring unit ₆₁ corresponding to the platform S1, the measurement unit 6 ₁₆ corresponding to the platform S16 of specifying the output point I2, specify the output point I3 of the filter to the measurement unit 6 ₁₅ corresponding to the platform S15, similar to the following, the filters for the measuring unit 6 ₁₁ corresponding to the platform S11 Specify the output point I7.

Also, the system state P2, as shown in FIG. 8, the filter specifies the output point I1 of the filter with respect to the measuring unit 6 ₂ corresponding to the load platform S2, the measuring unit ₆₁ corresponding to the platform S1 Specifies the output point I2 of specifying the output point I3 of the filter to the measurement unit 6 ₁₆ corresponding to the platform S16, hereinafter the same, the output of the filter to the measurement unit 6 ₁₂ corresponding to the platform S12 Point I7 is designated.

  In FIG. 9, in each of the system states P1 to P16, in order to determine whether or not to drive a gate opening / closing cylinder for distributing an object to be measured, a distribution output memory in which a distribution signal is held should be verified. Indicates a stand.

  In each of the system states P1 to P16, the stage on which the distribution output memory is to be verified is the stage that approaches the distribution positions (1) to (8) in FIG.

  Therefore, for example, in the system state P1, the platforms S16, S15, S14, S13, S12, S11. S10. S9 is each distribution position (1), (2), (3), (4). (5), (6). Since (7) and (8) are reached, these platforms S16, S15, S14, S13, S12, S11. S10. S9 is a platform to be verified.

  In the system state P2, the platforms S1, S16, S15, S14, S13, S12, S11. S10 is each distribution position (1), (2), (3), (4). (5), (6). Since (7) and (8) are reached, these platforms S1, S16, S15, S14, S13, S12, S11. S10 is the platform to be verified.

  Similarly, as the system state progresses, the stage on which the distribution output memory is to be verified is shifted by one.

  10A and 10B are flowcharts for explaining the operation of the present invention, and are executed by the arithmetic control circuit 18 of the centralized control unit 11. The built-in clock generation circuit interrupts the CPU of the arithmetic control circuit 18 with, for example, a 1 msec clock pulse, and is processed with the highest priority. That is, it is executed with the highest priority every 1 msec. The slits 33 and 34 of the annular ring 35 are formed so that each of the pulses Tp and Rp has a pulse width sufficiently longer than 1 msec even when the rotation of the turntable 2 is fastest.

  First, as shown in FIG. 10A, it is determined whether or not the timing pulse Tp is at a high level (step n1). If it is not at a high level, the system state transition flag Fc is set to “0” because it is not the system state transition timing. To step n8 (step n17).

  In step n1, when the timing pulse Tp is at a high level, it is determined that the system state transition timing is “0” (step n2), and the flag Fc is “0”. When it is "0", the process proceeds to step n3, the system state transition flag Fc is set to "1", and it is determined whether or not the reset pulse Rp is at high level (step n4). Assuming that it is the transition timing to the state P1, "1" indicating the system state P1 is set in the counter Px indicating which system state it is, and the process proceeds to step n6 (step n5).

  In step n4, when the reset pulse Rp is not at the high level, “1” is added to the count value of the counter Px and the process proceeds to step n6, assuming that it is a transition timing to a system state other than the system state P1. Step n18).

  In step n6, "1" is set to the system state start flag Fs indicating that the system state of the count value of the counter Px has started, and the process proceeds to step n7.

  In step n7, the distribution output memory of a predetermined platform corresponding to the value of the system state Px shown in FIG. 9 is verified. For example, when the system state is P1, as shown in FIG. 9, the distribution output memories of the platforms S16 to S9 approaching the distribution positions (1) to (8) are verified. If the distribution output memory is “1”, a drive signal for the gate opening / closing cylinder is set, and the process proceeds to Step n9.

  This step n7 is processing corresponding to the above-mentioned time point (4) in FIG. 6, and details thereof are shown in FIG.

  As shown in FIG. 11, according to the value of the system state Px, a distribution output memory for a predetermined platform is tested, and a process for driving a gate opening / closing cylinder is performed.

For example, when the system state Px = 1 (system state P1), it is determined whether or not the distribution output memory MR16 of the mounting table S16 is “1” (step n). The drive flag FV16 for driving the cylinder for opening and closing the gate of the table S16 is set to “1” (step n7 ₁ −1), the distribution output memory MR16 is reset to “0”, and the process proceeds to step n7 ₁ -4. Move (step n7 ₁ -3).

In step n7 ₁ -4, it is determined whether the sorting output memory MR15 of the platform S15 is "1", when it is "1", for driving the cylinder for gating the platform S15 It sets the drive flag FV15 to "1" (step n7 ₁ -5), and resets the sorting output memory MR15 to "0" and proceeds to the next step (step n7 ₁ -6).

  Similarly, the distribution output memories MR14, S13, S12, S11, and S10 of the platforms S14, S13, S12, S11, and S10 are verified, and processing is performed for driving the gate opening / closing cylinder.

In step n7 ₁ -21, it is determined whether the sorting output memory MR9 of the platform S9, "1", when it is "1", drives the cylinder for gating the platform S9 "1" is set in the drive flag FV9 for (step n7 ₁ -22), the flow proceeds to step n9 in Fig. 10 resets the sorting output memory MR9 to "0" (step n7 ₁ -23).

Further, when the system state Px = 2, it is determined whether or not the distribution output memory MR1 of the mounting table S1 is “1” (step n7 ₂ −1), and when it is “1”, the mounting table S1. The drive flag FV1 for driving the gate opening / closing cylinder is set to “1” (step n7 ₂ -2), the distribution output memory MR1 is reset to “0”, and the process proceeds to step n7 ₂ -4 ( Step n7 ₂ -3).

In step n7 ₂ -4, it is determined whether or not the distribution output memory MR16 of the mounting table S16 is “1”. If it is “1”, the gate opening / closing cylinder of the mounting table S16 is driven. The drive flag FV16 is set to “1” (step n7 ₂ −5), the distribution output memory MR16 is reset to “0”, and the process proceeds to the next step (step n7 ₂ −5).

  In the same manner, the distribution output memories MR15, S14, S13, S12, and S11 of the platforms S15, S14, S13, S12, and S11 are verified, and processing is performed for driving the gate opening / closing cylinder.

In step n7 ₂ -21, it is determined whether or not the distribution output memory MR10 of the mounting table S10 is “1”. If it is “1”, the gate opening / closing cylinder of the mounting table S10 is driven. "1" is set in the drive flag FV10 for (step n7 ₂ -22), the flow proceeds to step n8 in FIG. 10 resets the sorting output memory MR10 to "0" (step n7 ₂ -23).

Similarly, according to each of the system states P3 to P15, the distribution output memory of the predetermined platform shown in FIG. 9 is verified, and the process for driving the gate opening / closing cylinder is performed, and the system state Px = 16 When it is, it is determined whether or not the distribution output memory MR15 of the mounting table S15 is “1” (step n7 ₁₆ −1). When it is “1”, the gate opening / closing cylinder of the mounting table S15 is turned on. The drive flag FV15 for driving is set to “1” (step n7 ₁₆ -2), the distribution output memory MR15 is reset to “0”, and the process proceeds to step n7 ₁₆ −4 (step n7 ₁₆ −3).

In step n7 ₁₆ -4, it is determined whether or not the distribution output memory MR14 of the mounting table S14 is “1”. If it is “1”, the gate opening / closing cylinder of the mounting table S14 is driven. The drive flag FV14 is set to “1” (step n7 ₁₆ −5), the distribution output memory MR14 is reset to “0”, and the process proceeds to the next step (step n7 ₁₆ −6).

  Similarly, the distribution output memories MR13, S12, S11, S10, and S9 of the platforms S13, S12, S11, S10, and S9 are verified, and processing for driving the gate opening / closing cylinder is performed.

In step n7 ₁₆ -21, it is determined whether the distribution output memory MR8 of the mounting table S8 is "1". If it is "1", the gate opening / closing cylinder of the mounting table S8 is driven. The drive flag FV8 is set to “1” (step n7 ₁₆ −22), the distribution output memory MR8 is reset to “0”, and the process proceeds to step n8 in FIG. 10A (step n7 ₁₆ −23).

  In this way, step n7 in FIG. 10A is performed, and the process proceeds to step n8 in FIG. 10A.

  In this step n8, it is determined whether or not the system state start flag Fs is “1”. If it is not “1”, the process proceeds to step n19 in FIG. 10B. When the system state start flag Fs is “1”, it is determined that the system state is started, and “1” is added to the counter Co for measuring the first and second predetermined times T01 and T0 (step n9). ), Go to Step n10.

  The first predetermined time T01 corresponds to the time from when the system state starts until the platforms S1 to S16 reach the position of the virtual extension lines I1, I2,... I7 in FIG. Waiting for T01 to elapse, the output point of the filter for obtaining the weight value is designated for the measurement unit on the predetermined stage. The second predetermined time T0 corresponds to the time required for receiving the weight value from the measurement unit specifying the output point of the filter and making it readable, and the system state is the same as the first predetermined time T01. Measurement starts after it starts.

  In step n10, it is determined whether or not a first predetermined time elapsed flag F01 indicating that the first predetermined time T01 has elapsed is "0". If it is not "0", the first predetermined time T01 has already elapsed. The process proceeds to step n14, and when it is “0”, the process proceeds to step n11, where it is determined whether or not the count value of the counter Co has reached the first predetermined time T01, and when it is not the predetermined time T01, the process proceeds to step n14. Move.

  When the count value of the counter Co reaches the first predetermined time T01 in step n11, the first predetermined time elapsed flag F01 is set to “1” (step n12), and the process proceeds to step n13.

In step n13, an output command for designating a stage with a predetermined number corresponding to the system state Px shown in FIG. 8 and a filter output point is set in the register, and an output command flag F indicating that the output command has been set. “1” is set to ₀ , and the process proceeds to Step n14.

This step n13 is processing corresponding to the above-mentioned time point (1) in FIG. 6, and an output command for acquiring a weight value from a predetermined filter output point of a predetermined stage shown in FIG. 8 according to the system state. To the measurement unit. For example, when a system state P1, as shown in FIG. 8, the filter output point I1 of the filter with respect to the measuring unit ₆₁ corresponding to the platform S1, the measurement unit 6 ₁₆ corresponding to the platform S16 of the output point I2, the output point I3 of the filter to the measurement unit 6 ₁₅ corresponding to the platform S15, similarly to below, the output point I7 of the filter to the measurement unit 6 ₁₁ corresponding to the platform S11 Send an output command to get the weight value by specifying each. Reading of the weight value transmitted from the measurement units 6 ₁ , 6 _{16 to} 6 ₁₁ in response to the output command will be described later.

  In step n14, it is determined whether or not the count value of the counter Co has reached the second predetermined time T0. If not, the process proceeds to step n19 in FIG. 10B.

 When the count value of the counter Co reaches the second predetermined time T0, the counter Co is reset to “0”, and the system state start flag Fs and the first predetermined time elapsed flag F01 are reset to “0” (step n15 ), “1” is set to the second predetermined time elapsed flag Fr indicating that the second predetermined time T0 has elapsed, and the process proceeds to step n19 in FIG. 10B (step n16).

  In step n19 of FIG. 10B, it is determined whether or not the driving flag FV1 for driving the gate opening / closing cylinder of the mounting table S1 is “1”. The cylinder for driving is started to open the gate, and "1" is added to the counter C1 to measure the operating time T1 of the cylinder (step n20), and the count value of the counter C1 becomes the operating time T1. (Step n21), and when the operating time T1 is not reached, the process proceeds to step n23. When the operation time T1 is reached, the count value of the counter is reset on the assumption that the opening and closing of the gate is completed, and the drive flag FV1 of the mounting table S1 is reset to “0” and the process proceeds to step n23 (step n22).

  In step n23, it is determined whether or not the drive flag FV2 for driving the gate opening / closing cylinder of the mounting table S2 is "1". If it is "1", the gate opening / closing cylinder of the mounting table S2 is determined. Is started to open the gate, and "1" is added to the counter C2 to measure the operation time T1 (step n24), and it is determined whether or not the count value of the counter C2 has reached the operation time T1. However (step n25), when the operating time T1 is not reached, the process proceeds to step n27. When the operation time T1 is reached, the count value of the counter is reset on the assumption that the opening and closing of the gate is completed, and the drive flag FV2 of the mounting table S2 is reset to “0” and the process proceeds to step n27 (step n26).

  Thereafter, similarly, the same processing is performed for the platforms S3 to S15. In step n28, it is determined whether or not the drive flag FV16 for driving the gate opening / closing cylinder of the platform S16 is "1". When it is “1”, the gate opening / closing cylinder of the mounting table S16 is driven to start opening the gate, and “1” is added to the counter C16 in order to measure the operation time T1 (step n29). Then, it is determined whether or not the count value of the counter C16 has reached the operating time T1 (step n30). When the operation time T1 is reached, the count value of the counter C16 is reset because the gate opening / closing is ended, and the drive flag FV16 of the mounting table S16 is reset to “0” and the process ends (step n31).

FIG. 12 is a flowchart showing the process of acquiring the weight value between the central control unit 11 and each of the measurement units 6 _{1 to} 6 _16, and this processing program is a program having a lower priority than the above-described FIG. is there.

  As shown in FIG. 12A, it is determined whether or not the output command flag F0 indicating that the output command for weight acquisition is set is “1” (step n101), and is “1”. In some cases, the output instruction flag F0 is reset to “0” (step n102), and the data output set in the predetermined output instruction register in accordance with the value of the current system state Px as shown in FIG. The command code, mounting number, and filter output point data sets are sequentially set in the serial controller 12 and transmitted to the corresponding measurement units, and the process ends (step n103).

 The measurement unit corresponding to each mounting number that receives these output commands calculates the average value up to the output point of the specified filter as the weight value, and sets the data content code, mounting number, and weight value for each It is set in the serial controller 10 and transmitted to the serial controller 12 of the centralized control unit 11.

As shown in FIG. 12B, the central control unit 11 determines whether or not there is a read request from the serial controller 12 (step n201). When there is a read request, the central control unit 11 determines a predetermined number corresponding to the mounting number k. The weight of the filter output transmitted from the measurement units 6 _{1 to} 6 ₁₆ is stored in the weight value register (WMRk) k = _{1 to 16,} and the process is terminated (step n202).

  The processing shown in FIG. 12 is performed at the above-mentioned second predetermined time T0 msec after transmitting an output command signal to each platform, that is, at the second predetermined time T0 msec shown in step n14 of FIG. 6 and FIG. 10A. Executed.

  After the second predetermined time T0 elapses, processing such as weight rank determination and distribution output memory update is performed based on the acquired weight value before shifting to the next system state.

  FIG. 13 shows an outline of this processing, and is a program having a lower priority than the processing program of FIG. 10 described above.

As shown in FIG. 13, whether or not the second predetermined time T0 is completed, that is, whether or not the above-mentioned second predetermined time elapsed flag Fr indicating that the second predetermined time T0 has elapsed is “1”. (Step n301), when the second predetermined time T0 has elapsed, the second predetermined time elapsed flag Fr is reset to “0” (step n302), and PG1 to PG16 are set according to the value of the system state Px. selecting and executing one of the programs of the (step n304 ₁ ~304 _16).

Then, if a system state Px = 1 (system state P1), i.e., the steps N304 ₁ of PG1 processing details will be described with reference to FIG. 14A~ Figure 14F.

  Since the system state P1 is present at the time of execution of this program, the weight values are stored in the weight value registers WMR1, WMR16-11 corresponding to the platforms S1, S16-S11 as shown in FIG.

  As shown in FIG. 14A, the weight value of the mounting table S1 is read by the weight value register WMR1 (step n401). Next, the weight value (internal count level) Wn1 of the object to be weighed on the mounting table S1 is calculated according to the following equation (step n402).

Wn1 = K1. (Wa1-Wi1) -Wz1
Here, K1 is the span coefficient of the stage S1, Wa1 is the weight value stored in the weight value register WMR1, Wi1 is the initial load of the stage S1, and Wz1 is the zero point load of the stage S1.

  Next, the calculated weight value Wn1 of the weighing object is converted into the weight value Wd1 of the display value level (step n403), the weight rank is determined based on the boundary weight value of FIG. 2 (step n404), and step n405. Move on.

  In step n405, it is determined whether or not the determined weight rank is the weight rank (1). If it is the weight rank (1), the weight rank (1) is determined, and the distribution position ( In order to distribute to 1), “1” is set in the distribution output memory MR1 for the mounting table S1 (step n406). Next, the weighing completion sign of the mounting table S1 is set (step n407), the weight value of the mounting table S1 is confirmed (step n408), and the process proceeds to step n409 in FIG. 14B. The weight value determined in step n408 is used for display, output, and tabulation.

  If it is not the weight rank (1) in step n405, the weight value acquired in the next system state P2, that is, the weight value is acquired at the next weight acquisition position and the weight rank is determined, so the weight value is determined. Without proceeding to step n409 of FIG. 14B.

  In step n409 of FIG. 14B, it is determined whether or not there is a weighing completion sign for the mounting table S16. If there is no weighing completion sign, the weight value for the loading table S16 is read by the weight value register WMR16 (step n410).

  Next, the weight value (internal count level) Wn16 of the object to be weighed on the platform S16 is calculated, converted into the display value level weight value Wd16 (step n411), and the weight rank is determined (step n412). The process moves to step n413.

  In step n413, it is determined whether or not the determined weight rank is the weight rank (2). If the weight rank is (2), the weight of the mounting table S16 is determined (step n416). Next, the weighing completion sign of the mounting table S16 is set (step n417), and “1” is set in the distribution output memory MR16 for the mounting table S16 in order to distribute to the distribution position (2) in the next system state P2. Set (step n418), then proceed to step n419 in FIG. 14C.

  If the determined weight rank is not the weight rank (2) in step n413, it is determined whether or not the determined weight rank is the weight rank (1) (step n414). In step n413, when it is not the weight rank (2), the object to be weighed is the weight rank (3) or higher, or the weight in the vicinity of the boundary weight value Wc1 between the weight rank (1) and the weight rank (2). It is considered to have a weight rank (1). Therefore, in step n414, it is determined whether or not the weight rank (1), and if it is the weight rank (1), the distribution position (1) has already been passed, so the weight rank (2) is determined. At the same time, the weight value is fixed to the lower limit weight value Wc1 of the weight rank (2), and the process proceeds to Step n417 (Step n415). That is, the object to be weighed is the weight rank (1) having a weight in the vicinity of the boundary weight value Wc1 between the weight rank (1) and the weight rank (2), but has already passed the sorting position (1). In order to assign weight rank (2) to the distribution position (2) in the next system state P2, “1” is set in the distribution output memory MR16, and the weight value is set to the lower limit of the weight rank (2). The weight value Wc1 is determined so that an error as a weight measurement value is reduced.

  In step n414, when it is not the weight rank (1), it is equal to or higher than the weight rank (3). Therefore, the weight value acquired in the next system state P2, that is, the weight value is acquired at the next weight acquisition position. Since the determination is performed, the process proceeds to step n419 in FIG. 14C without determining the weight value.

  In step n419 of FIG. 14C, it is determined whether or not there is a weighing completion sign for the mounting table S15. If there is no weighing completion sign, the weight value for the loading table S15 is read from the weight value register WMR15 (step n420). Next, the weight value (internal count level) Wn15 of the object to be weighed on the platform S15 is calculated, converted to the weight value Wd15 of the display value level (step n421), the weight rank is determined (step n422), and the step Move to n423.

  In step n423, it is determined whether or not the determined weight rank is the weight rank (3). If the weight rank is (3), the weight of the mounting table S15 is determined (step n426). Next, the weighing completion sign of the mounting table S15 is set (step n427), and “1” is set in the distribution output memory MR15 for the mounting table S15 in order to distribute to the distribution position (3) in the next system state P2. Set (step n428), then proceed to step n429 in FIG. 14D.

  If the determined weight rank is not the weight rank (3) in step n423, it is determined whether or not the determined weight rank is the weight rank (2) (step n424) and is the weight rank (2). Sometimes, the object to be weighed is weight rank 2 having a weight in the vicinity of the boundary weight value Wc2 between the weight rank (2) and the weight rank (3), but has already passed the sorting position (2). Rank (3) is determined, and the weight value is determined to be the lower limit weight value Wc2 of weight rank (3), and the process proceeds to step n427 (step n425).

  In step n424, when it is not the weight rank (2), it is equal to or higher than the weight rank (4). Therefore, the weight rank is determined based on the weight value acquired in the next system state P2. Move to step n429 of 14D.

  In step n429 of FIG. 14D, the same processing is performed for the mounting table S14. In step n430, the same processing is performed for the mounting table S13. Further, in step n431, the same processing is performed for the mounting table S12. The process moves to step n432.

  In step n432 of FIG. 14E, it is determined whether or not there is a weighing completion sign for the mounting table S11. If there is no weighing completion sign, the weight value for the loading table S11 is read from the weight value register WMR11 (step n433). Next, the weight value (internal count level) Wn11 of the object to be weighed on the platform S11 is calculated, converted to the weight value Wd11 of the display value level (step n434), the weight rank is determined (step n435), and step Move to n436.

  In step n436, it is determined whether or not the determined weight rank is the weight rank (7). If the weight rank is the weight rank (7), the weight of the platform S11 is determined (step n440). Next, the weighing completion sign of the mounting table S11 is set (step n441), and in order to distribute to the distribution position (7) in the next system state P2, “1” is set in the distribution output memory MR11 for the mounting table S11. Set (step n442), then proceed to step n443 in FIG. 14F.

  If the determined weight rank is not the weight rank (7) in step n436, it is determined whether or not the determined weight rank is the weight rank (6) (step n437), and is the weight rank (6). Sometimes, the object to be weighed is the weight rank (6) having a weight in the vicinity of the boundary weight value Wc6 between the weight rank (6) and the weight rank (7), but has already passed the distribution position (6). The weight rank (7) is determined, and the weight value is fixed to the lower limit weight value Wc6 of the weight rank (7), and the process proceeds to Step n441 (Step n438).

  In step n437, when the determined weight rank is not the weight rank (6), it is the weight rank (8). Therefore, the weight value of the platform S11 is determined, and this weight value is stored in the temporary storage memory. The system state P2 is transferred to step n443 in FIG. 14F (step n439).

  In step n443 of FIG. 14F, it is determined whether or not there is a weighing completion sign of the mounting table S10. If there is a weighing completion sign, the process proceeds to step n447. When there is no weighing completion sign, the determined weight value of the mounting table S10 stored in the previous system state P16 is read from the temporary storage memory to determine the weight value of the mounting table 10 (step n444). Next, the weighing completion sign of the mounting table S10 is set (step n445), and “1” is set in the distribution output memory MR10 for the mounting table S10 in order to distribute to the distribution position (8) in the next system state P2. The setting is completed (step n446), and the measurement completion sign of the stage S9 that moves outside the discharge range in the next system state is reset and the process ends (step n447).

  As described above, in this system state P1, the weights are read as shown in FIGS. 14A to 14E for the platforms S1 and S16 to S11 corresponding to the system state P1 shown in FIG. The stage S10 is processed according to the weight value stored in the previous system state P16 as shown in FIG. 14F.

  In the next system state P2, weights are read and processed for the platforms S2, S1, S16 to S12 corresponding to the system state P2 shown in FIG. 8 as in FIGS. 14A to 14E. In the same manner as in FIG. 14F, the processing is performed according to the weight value stored in the previous system state P1.

  In the same manner, as the system state advances, the stage is shifted one by one as shown in FIG. 8, and the same processing as in FIGS. 14A to 14F is performed.

In this embodiment, as the filter provided in the arithmetic circuit 9 of the measuring unit ₆₁ through _16, have been described in the average filter is a FIR filter cascade (slave) connection, the filter type is not limited thereto.

For example, as an IIR type filter, as shown in FIG. 15B, a first-order lag analog low-pass filter as shown in FIG.
A = (CR / t) / [(CR / t) +1]
B = 1 / [(CR / t) +1]
t: Sampling time interval is converted into a digital filter, these are cascade-connected as shown in FIG. 15C, and the second average value arithmetic circuit (AV2) 31 for attenuating the natural frequency of the stage shown in FIG. Is input to this filter row every 10 msec, and the output values I1, I2,... I7 are calculated as shown in the calculation block of FIG. 1 may be read in accordance with the positions of the virtual extension lines I1, I2,... I7 shown in FIG.

In FIG. 15, Z ⁻¹ indicates a delay element that delays data by one sampling time.

  In this case, the constants A and B are converted into values indicating response characteristics inversely proportional to the magnitude of the rotational speed when the rotational speed is set. The values of A and B are automatically calculated and set as A1x, A2x, B1x, and B2x according to the set arbitrary rotation speeds. That is, the time constant is determined so that the response time of the filter falls within the above-described allowable response time according to the rotation speed.

  Note that a second or higher order filter may be used.

(Embodiment 2)
FIG. 16 is a diagram corresponding to FIG. 4 described above according to another embodiment of the present invention.

  In this embodiment, a filter is set for a vibration noise signal having a frequency corresponding to the rotation speed.

  As shown in FIG. 1, it is expected that the platforms S1 to S16 to which the object to be weighed 13 is supplied move from the position of the virtual extension line I0 to the position of the virtual extension line I1, that is, the maximum rotation speed. In the case of 750 msec, the response time of 1500 msec in the case of the minimum rotation speed (this is referred to as the allowable response time) is expected, and the vibration noise signal of a specific frequency that varies depending on the rotation speed among basic vibrations is smoothed. Provide a notch filter to increase the characteristics.

  The rotary table 2 including the platforms S1 to S16 of the weight sorting apparatus 1 shown in FIG. 1 receives a rotary column with a bearing and rotates by transmission of rotational power from a driving motor 19, but slides on the bearing. In response to vibration generated in the drive mechanism, this becomes basic vibration noise for the load signals of the platforms S1 to S16. The frequency of the fundamental vibration noise is proportional to the rotation speed.

  As shown in FIG. 16, a notch filter for the fundamental vibration noise is inserted between the output point I0 and the output point I1 in FIG. The second average value calculation circuit (AV2) 31 in FIG. 16 corresponds to the second average value calculation circuit (AV2) 31 in FIG.

  The response time of this type of filter is assigned to a part of the time (= weight measurement interval) required for the platforms S1 to S16 to move from the virtual extension line I0 to I1 shown in FIG. A notch filter for mounting base vibration for attenuating the basic vibration signal generated in the load signal according to the rotation of ~ S16 is set.

  This filter can be set by the following operation procedure, for example.

  In the adjustment stage, the platform is rotated at a predetermined speed, and frequency analysis of the load signal of the specific platform k is designated from the input unit 24 attached to the display setting unit 22. As a result, an A / D sampling load signal of one or more rotations of the designated platform (number k) is sent to the central control unit 11, and noise included in the load signal by the Fourier transform calculation means provided in the central control unit 11. Analyze the frequency of the signal.

  As a result of this frequency analysis, the frequency or period and amplitude of the noise signal having an amplitude of a predetermined level or higher are sent to the display setting unit 21 and displayed on the display unit 25 attached to the display setting unit 21. As a result, the period of the noise signal having an amplitude of a predetermined level or more generated in the load signal as the basic vibration signal as the mounting table rotates once is detected. The value of the period Tbx of the detected fundamental vibration noise signal having an amplitude of a predetermined level or more is set from the input unit 24.

  However, the magnitude of the amplitude can be determined by calculating the amplitude of the natural vibration signal, which is a transient response vibration signal of the platform generated when the object to be weighed is dropped on the platform and supplied from the supply conveyor, by the above fast Fourier transform calculation. Seek by means. In particular, when the rotational speed is the highest, the object to be weighed is supplied to the platform, and the amplitude value when the platform reaches the position of the virtual extension line I1 is obtained and compared using the amplitude as a reference. This facilitates determination of the magnitude of the influence on the weight value.

  The relationship between the period Tbx of the basic vibration noise signal and the rotation speed V at the time of generation of the vibration noise signal is expressed as Tbx = K · (1 / V), where the rotation speed V and the basic vibration signal generated along with the rotation A coefficient K relating to one period Tbx is determined. By setting the coefficient K in advance in the adjustment stage, the period Tbx of the basic vibration noise signal generated at the rotation speed V is automatically calculated by setting the rotation speed V during the first operation.

  If there are a plurality of vibration noise signals whose frequency is proportional to the rotation speed and whose amplitude level cannot be ignored in terms of measurement accuracy, the same filters are cascaded.

The rotation speed may be set as a weighing processing capacity value, that is, a numerical value in units of pieces / minute. If the number U of weight sorters is determined, Q / min is set,
V = Q / U
What is necessary is just to convert into the value of the rotational speed V by the relationship.

When the rotation speed is set to Vrpm,
(15000 / V) −Tbx = (15000 / V) −K · (1 / V)> 0
If so, the response time of the notch filter falls within the time required for the platform to move from the position of the virtual extension line I0 to the position of the virtual extension line I1, so that setting is possible.

  As shown in FIG. 16, since the data output from the second average value calculation circuit (AV2) 31 is 1 msec as a filter for the basic vibration signal having the period Tbxmsec calculated above, the period of the basic vibration signal is A shift register SFRb composed of Tbx cell registers, which is the number equal to the value Tbx, is prepared.

  The response time by the shift register SFRb is Tbxmsec.

  The data of each cell register of the shift register SFRb is shifted right every 1 msec, the output of the second average value arithmetic circuit (AV2) 31 is inputted every 1 msec, and the output of each cell register of the shift register SFRb is changed every 1 msec. 3 The average calculation is performed by the average value calculation circuit (AV3) 38.

  By this calculation operation, the fundamental vibration signal having a cycle of Tbx is particularly effectively removed from the data of the output point of the third average value calculation circuit (AV3) 38.

  Since the response time was delayed by Tbxmsec by adding a new filter in the response time expected for the four sections I0 to I1 of the cell register, the response time from the output points I0 'to I1 is the maximum rotation speed. In some cases, (750-Tbx) msec, and in the case of the minimum rotation speed, (1500-Tbx) msec.

In the case of an arbitrary rotation speed Vrpm, from the above, it can be estimated that the movement time (15000 / V) sec of the four sections from the output points I0 to I1 as the allowable response time, so that the remaining response time after subtracting Tbx is
(15000 / V) −Tbx = (15000 / V) −K · (1 / V) (msec)
Can be expected.

  Here, the average value by the third average value calculation circuit (AV3) 38 is sent to the third shift register SFR3 at a time interval of 10 msec. The number of cell registers provided from the output point I0 ′ to the output point I1 of the third shift register SFR3 is (750−Tbx) / 10 in the case of the maximum rotation speed, and (1500−Tbx) / in the case of the minimum rotation speed. Only 10 are prepared in the memory. At an arbitrary rotation speed Vrpm, only [(15000 / V) −K · (1 / V)] / 10 are prepared.

  When the output point I1 is specified, the average value of the outputs of [(15000 / V) −K · (1 / V)] / 10 cell registers is calculated by the fourth average value arithmetic circuit (AV4) 39. And output as a weight acquisition value.

  When the output points I2, I3,... Are specified, the average value of the cell register output between the output points I0 'to I2, I3,. Calculated at 39 and output as a weight acquisition value

  If the response time of the inserted filter is assigned to the basic vibration signal generated by the rotation and there is still a margin for the allowable response time until the weight value is acquired, the weight value at the I1 position is further increased. In order to obtain a stable value, the response time is [(15000 / V) −K · (1 / V)] msec, and an average value filter calculated by the fourth average value calculation circuit (AV4) 39 is added. is there.

  Since the result of the average calculation by the fourth average value calculation circuit (AV4) 39 does not affect the result calculated in the preceding stage, this filter is a subordinate connection (cascade connection). The longer the response time can be set, the better the smoothing characteristic from the impulse vibration to the long period vibration.

  The constant of the filter configured in the time zone in which the platform moves from the output point I0 to the position of I1 is determined by the distance from the supply position of the object to be measured to the position where the weight value is obtained and the rotational speed of the platform.

  In this embodiment, a filter having a smoothing characteristic suitable for a fundamental vibration signal whose period changes depending on the magnitude of the rotation speed in the allowable response time that varies depending on the above is provided. Thereby, noise due to vibration of the turntable 2 can be effectively removed.

  Other configurations are the same as those of the above-described embodiment.

(Other embodiments)
(I) In the above description, as a preferred method, the filter provided in the measurement unit has a smoothness as the weight value acquisition position moves away from the supply position of the object to be weighed along the rotation direction of the platform, that is, backward. Is configured to produce a large output.

  However, the weight value may be acquired from the output of the same type of filter at any weight value acquisition position. In this case, as the distance from the point at which the object to be weighed is supplied to the platform increases, the transient response signal due to the object to be weighed on the platform converges smaller toward the stationary weight value, so only one type of filter is used. Even in the case where the weight value is used, the weighing accuracy can be improved only by acquiring the weight value by changing the acquisition timing of the output load signal.

(Ii) In addition, the point where the acquisition position of the weight value is provided corresponding to the distribution position,
It is not always necessary to provide them one by one in front of all the sorting positions. If the weight is measured at least two times from the conventional method of measuring the weight at only one place, the total weight value obtained (= measurement) Contributes to the improvement of weighing accuracy.

  For example, for the weight ranks (1) to (4), the weight rank is determined using the weight value acquired at the position of the imaginary line I1 in FIG. 1, and for the weight ranks (5) to (8), FIG. It may be determined using the weight value acquired at the position of the virtual line I5.

(Iii) In the above-described embodiment, N platforms are arranged at positions where the circumference is divided into N at equal intervals, and the weight of the object to be weighed placed on the platform is set according to the size of the weight. In the most extreme case, a rotary weight sorter that sorts the weight according to the determined weight rank is given to one of a plurality of sorting positions provided on the circumference. In the N sections, there is only one section provided with a platform, and the platform reaches the weight of the object to be weighed supplied to one platform before each sorting position. The apparatus may be such that a weight value is newly acquired each time a weight rank is determined, and a weight rank distribution position corresponding to the determined weight rank is allocated.

(Iv) In the embodiment shown in FIG. 1, the distribution positions are provided so that the weight rank (1) to (8) is increased from the smallest weight rank to the largest weight rank in the rotation direction of the mounting table on the circumference. It was. However, according to the convenience of the user, the sorting positions are not always arranged from the one with the smallest weight rank to the one with the largest weight rank. In the present invention, the weight ranks may be arranged in any order.

  For example, the weight rank distribution position is arranged so as to correspond to the weight rank (3) → (5) → (4) → (6) → (2) → …… along the rotation direction of the mounting table on the circumference. Suppose that

  In this case, when a rank other than the weight rank (3) is determined based on the weight value acquired at the position of the virtual extension line I1 in FIG. 1, the weight is newly calculated based on the weight value acquired at the weight acquisition position after the virtual extension line I1. The rank is determined.

  Now, (a) the true weight value of the object to be weighed is in the vicinity of Wc2, which is the boundary weight value between the weight rank (2) and the weight rank (3), and the true weight value belongs to the weight rank (3). And (b) the true weight value of the object to be weighed is in the vicinity of Wc3 which is the boundary weight value between the weight rank (3) and the weight rank (4), and the true weight value is the weight rank (3). Consider the case of belonging to

  Even if the weight of the object to be weighed acquired at the position of the virtual extension line I1 is (a) based on the noise signal included in the load signal, the weight rank (2) is determined, and in the case of (b). However, the weight rank (4) is determined, and as a result, the weight rank (3) is not sorted and may pass.

  This object cannot occur until the weight rank (5) is determined based on the weight value acquired at the position of the next virtual extension line I2, but it is also determined by the weight value acquired at the position of the virtual extension line I2. If it is an object to be weighed in the case of (a), it is in either the weight rank (2) or the weight rank (3), and if it is an object to be weighed in the case of (b), it is weight rank (3). It is determined as one of the weight ranks (4).

  However, since the weight rank for distribution at this timing is (5), the weight rank (5) is not determined regardless of whether the object to be weighed is (a) or (b). It does not coincide with the position (5), and the object to be weighed passes without being distributed.

  Next, if the object to be weighed is (a) based on the weight value acquired at the position of the virtual extension line I3, the weight rank (2) or the weight is the same as the weight value acquired at the position of the virtual extension line I2. If it is an item to be weighed in any of ranks (3) and (b), it is judged as either weight rank (3) or weight rank (4). When it is a weighing object, the distribution position at this timing is (4), and if it is determined as the weight rank (4), it can be distributed as it is, but it is determined as the weight rank (3) that has already passed. If it is, the weight rank is forcibly determined as the weight rank (4), and the weight value is forcibly set to the lower limit value Wc3 of the weight rank (4).

  On the other hand, if the object to be weighed is (a) based on the weight value acquired at the position of the virtual extension line I5, the weight rank is determined as either the weight rank (2) or the weight rank (3), and the weight If it is a rank (2) determination, it can be assigned as it is, but if it is determined to be a weight rank (3) that has already passed, the weight rank is forcibly determined as the weight rank (2), and the weight The value is also forcibly set to a weight value (Wc2−minimum display amount) obtained by subtracting the minimum display amount from the boundary weight value Wc2, which is the upper limit weight value of the weight rank (2). Since the boundary weight value Wc2 belongs to the weight rank (3), the weight value obtained by subtracting the minimum display amount is set as the upper limit weight value of the weight rank (2).

  In the conventional example, when the weight of the object to be weighed is (b), the weight value acquired at the position of the virtual extension line I1 still contains a large noise signal, and the true weight is in the weight rank (3). If a weight rank (4) is determined by acquiring a larger weight value even if it belongs, the weight value (including a larger noise signal) and the weight rank are determined according to the determination. It was.

  On the other hand, in this embodiment, since the initial distribution position is (3), the distribution is not performed, and when the weight rank (3) is determined by the more accurate weight value acquired at the position of the virtual extension line I3. Since the distribution position of the weight rank (3) has already passed, it cannot be determined as the weight rank (3) and is set to the weight rank (4), but the weight value is determined as the lower limit value of the weight rank (4). .

  This means that the weight rank is different from the correct weight rank (3), but the weight value is at least closer to the true weight value of the object to be weighed than in the conventional example, and the weight value determined in the conventional example. There is an effect that a higher accuracy can be obtained.

  That is, when the weight rank is determined based on the acquired weight value, if the determined weight rank has already passed, the acquired weight value is within the two weight ranks adjacent to the upper and lower sides of the determined weight rank. The weight value is determined to be the weight rank of which the boundary weight value is closer, and the weight value is the lower limit value of the weight rank when the determined weight rank is larger, and the upper limit of the weight rank when the weight weight is smaller. The value is forcibly determined together with the weight rank judgment value.

  That is, when the weight rank is determined based on the acquired weight value, if the determined weight rank has already passed, the acquired weight value and the one heavy weight rank adjacent to the weight rank determined based on the weight value. The absolute value of the difference between the lower limit boundary value and the upper limit boundary value of one light weight rank may be calculated and compared, and the weight rank with the smaller absolute value may be determined again.

1 rotary weight sorting device 2 turntable 4 supply conveyor 5 _{1-5 16} load sensor ₆₁ through ₁₆ measurement unit 9 arithmetic circuit 11 central control unit 21 the display setting unit 24 input unit 25 display unit S1~S16 the platform

Claims (5)

Based on a driving means for rotating the measuring instrument along a circumference around the center of rotation, and a load signal from the measuring instrument supplied with the measuring object at a supply position on the circumference, the measuring object And determining means for determining a weight rank to which the object to be weighed belongs among a plurality of weight ranks ,
A plurality of distribution locations corresponding individually to said plurality of weight classes are arranged along the circumference, of the distribution positions of the multiple, the distribution position corresponding to the weight rank is determined by said determining means , A rotary weight sorting device for sorting the objects to be weighed,
Said plurality of distributing positions, along said direction of rotation of the measuring instrument to be weighed is fed by the supply position, the supply position as the upstream, are arranged toward the downstream side of the supply position,
The determination means includes the load signal when the weighing instrument to which the object to be weighed is supplied has reached a first measurement position upstream of the plurality of sorting positions arranged along the circumference. The weight rank is determined by measuring the weight of the object to be weighed, and further, the weighing machine to which the object to be weighed is distributed at the plurality of distribution positions until the object to be weighed is distributed, When one or more measurement positions are reached, the weight rank is determined by measuring the weight of the object to be weighed based on the load signal at that time .
A rotary weight sorter characterized by that. A plurality of the measuring instruments;
The distribution position corresponding to the weight rank determined by the weight of the object measured at the first measurement position is one or more reachable before the object to be weighed is distributed to the downstream weighing instrument. the number of measurement positions is increased,
The rotary weight sorter according to claim 1. Wherein the first measurement position, measuring instruments objects to be weighed are supplied, among the plurality of distributing positions, a position immediately before reaching the first sorting position on the most upstream side, front SL one or more measurement positions Corresponds to a distribution position downstream of the first distribution position, and is a position immediately before reaching the distribution position,
The determination means determines the weight rank by measuring the weight of the object to be weighed based on the load signal from the weighing instrument every time it reaches each measurement position immediately before each distribution position.
The rotary weight sorter according to claim 2. When the weight rank determined based on the measured weight of the weighing object is a weight rank corresponding to the distribution position that the weighing instrument has already passed, based on the measured weight of the weighing object. , Re-determine the weight rank one lighter or heavier than the determined weight rank,
The rotary weight sorter according to claim 3. Distributing the objects to be weighed in the measuring device to the distribution position corresponding to the weight rank re-determined by the determining means;
The rotary weight sorter according to claim 4.

Images