JPS5950318A - Dynamic weighing method - Google Patents

Abstract

PURPOSE:To perform dynamic weighing with high accuracy even if the load distributed on a belt varies largely or pulleys are loaded considerably eccentrically by segmenting the overall length in one rotation of a conveyor to plural blocks, and calculating the differences between the zero point weight values in the respective blocks and the weight value in the block when conveying the articles to be weighed. CONSTITUTION:It is assumed that a signal for commanding measurement is held supplied to an AND gate 38 and that the gate is held open while a belt conveyor 1 conveys articles to be weighed. When a rotation detector 7 generates rotation pulses in this state, a counter 22 calls successively registers 20-1-20-m thereafter. When the articles to be weighed arrive on a balance part 2 and a detector 6 forms a detection signal, a subtractor 32 subtracts the value stored in the register called at that time from the A/D data from an A/D converter 4 and supplies the same to an arithmetic calculation circuit 34 for weight average. The subtractor 32 subtracts the pulse signal from said value at every generation of the pulse signal from a pulse generator 5 while the detector 6 makes a detection signal. The circuit 34 integrates the subtracted values, averages further the same, outputs the average value and displayes the value on a display part.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for dynamically weighing articles to be weighed by means of an abrasive conveyor.

In a weighing conveyor that combines a belt conveyor and a weighing device, the zero point fluctuates even in a no-load state due to the unevenness of the distributed load on the belt of the belt conveyor and the eccentric load of the pulley that drives the belt. There was a problem that it was not possible to perform highly accurate measurements. Therefore,
Conventionally, there has been a method of determining the zero point weight at appropriate n points in one rotation of a belt conveyor, and dynamically weighing by using the average value of these as the zero point weight value. However, even with this method, if there are large variations in the belt's distributed load or if the pulley's eccentric load is large, the difference between the average zero point weight value and the actual one-loss zero point weight value will become large, and the accuracy will still be affected. It was not possible to make a high measurement.

SUMMARY OF THE INVENTION An object of the present invention is to provide a highly accurate dynamic weighing method even when there are large variations in the distributed load on a belt or when there is a large eccentric load on a pulley.

Therefore, in the dynamic weighing method according to the present invention, the entire length of the conveyor in one revolution is divided into a plurality of sections sufficient to maintain the metering cleanliness, and the conveyor is rotated at least once in an unloaded state, and each section is Obtain the zero point weight values of
When conveying an article to be weighed on a conveyor, the weight value of the section in which the article is generally transported is obtained, and the difference between this weight value and the zero point weight value corresponding to this section is calculated. be.

Hereinafter, this invention will be explained in detail based on two embodiments. A first embodiment is shown in FIGS. 1 to 4. In Fig. 1, numeral 1 is a belt conveyor, 2 is a weighing section, 3 is a load cell, 4 is a sequential comparison type converter or double integral type A/D converter,
Reference numeral 5 denotes a pulse generator, which provides conversion timing to the A/D converter and also provides zero point detection timing to a control section, which will be described later. 6 is a detector for detecting that the article to be weighed has reached the weighing section 2; 7 is a rotation detector that generates a pulse every time the belt conveyor rotates once.

The outputs of the detector 6, rotation detector 7, pulse generator 5, and A/D converter 4 are input to a microcomputer 8. As shown in the figure, the microcomputer 8 includes a control section 9, a calculation section 10, and a register IINA/D timing pulse counter (hereinafter referred to as a timing counter) 12.
, belt rotation number counter (hereinafter referred to as CHC counter).

) 13 and a display unit 14. 15
is a keyboard that supplies start signals and the like to the microcomputer 8.

The microcomputer 8 is programmed according to the flowcharts shown in FIGS. 3 and 4 to carry out the dynamic weighing method according to the present invention. FIG. 3 shows the main routine, and FIG. 4 shows the interrupt routine. That is, when a start signal is input from the keyboard 15, the control section 9 determines whether or not the belt z of the conveyor 1 is being driven. Static weighing is performed when the machine is not in operation. Since this is irrelevant to the present application, the explanation will be omitted. If it is being driven, the control unit 9 clears the register 11, and then the CHC
Clear the counter 13 and set the dynamic zero measurement command flag to “1”
Make it. Then, it waits for the pulse generator 5 to generate a pulse signal, and when it occurs, jumps to the interrupt routine shown in FIG.

In the interrupt routine, when the rotation detector 7 generates a pulse, the control unit 9 determines whether the start pulse flag is "0" or "1".
Determine whether Since the start pulse flag is "0" here, the timing force counter 2 is cleared and the belt start flag is set to "l".

Next, determine whether the weight is being measured or not. Currently, the conveyor 1 is not conveying any articles to be weighed, and the bell 1-
,/Determine whether the start flag is "1" or not. Here, since the belt 7 start flag has already been set to "1", the CHC counter is set to "1". Then CH
It is determined whether the C counter 13 is O or not. Note that the above-mentioned determination of the dynamic zero measurement command flag, determination of the belt surprise start flag, and determination of whether or not the CHC counter 13 is zero are intended to prevent measurement from being affected even if the microcomputer malfunctions. be.

Next, it is determined by 't4J whether the count value of the CHC counter 13 is a value obtained by adding 1 to the predetermined number of revolutions N of the conveyor. Here, since the two are not equal, the data from the A/D converter 4 is read and stored in the register indicated by the count value of the timing counter 12. That is, in the operation up to now, the pulse generator 5 generates a pulse (the bell shown in FIG. 2(b)) only after the output of the load cell 3 shown in FIG. 2(a) is generated.
When the signal 16) of e) is generated, it is A/D converted and stored at address ○ in the register 11. After this, the third
Returning to the main routine shown in the figure, it is determined whether the dynamic weighing method flag is "1". Since it is not "1" here, when the timing pulse is input again, the process returns to the interrupt routine again.

In the interrupt routine, it is determined again whether there is a start pulse for belt 7, but since conveyor 1 has not yet completed one revolution, there is no start pulse, so the start pulse flag is set to "0" and the timing counter is incremented by one. That is, set it to "1", then set the belt start pulse flag to "0". Thereafter, in the same way as described above, it is determined whether the weight is being measured or not, it is determined whether the dynamic zero measurement command flag is "l", and then it is determined whether the belt no start flag is "1" or not. be done. Here it is "0" so CHC
The CHC counter 13 is set to ``0'' without incrementing the counter 13.
” or not. Thereafter, the A/D data is read into the register address 1 in the same manner as the first time, and the process returns to the main routine. Thereafter, A/D data is read in the same way when the next start pulse is input.

The number of A/D data obtained in this way is assumed to be m.

When the next start pulse is input, similarly, new A/D data is accumulated from addresses 0 to m-, 1 of the register 11, and the CHC counter 13 increments to "2". Thereafter, when the belt conveyor 1 makes N rotations in the same manner, the sum of N zero point measurement values for the same section of the conveyor is stored at address 0, and the same goes for other addresses. Then, when the belt conveyor 1 rotates by °N, -CHC counter 1
Since the count value of 3 becomes N-4-1, the CHC counter 13 is cleared, the dynamic zero measurement command flag is set to rOJ, the dynamic zero measurement end flag is set to "month," and the process returns to the main routine.

In the main routine, the dynamic-zero measurement end flag is “1”.
In this case, it is "1", so after setting the dynamic zero measurement command flag to "0", each integrated value stored in addresses 0 to m-1 of register 11 is divided by N. Find the average value and return each to its original address. Next, it is determined whether the measurement processing completion flag is rJ or not, but since it is "0" here, the next key interrupt routine is executed until it is determined whether weight measurement is in progress. Since the weight is being measured here, when the detection signal is input from the detector 6, the measurement processing command flag is set to rlJ, the A/D data is read, the average value zero point data specified by the address is read by the timing counter 13, and the A/D data is read. The average value zero point data is subtracted from the /D data, stored in the register 11, and the addition number counter is incremented by one. Then, it is determined whether the number of additions is 4 or not. the current,
Since this is not 4, the process returns to the main routine. That is, this means that the Noculus generator 5 converts the output of the output cell 3 shown in FIG. 2(C) into a pulse (see FIG. When 17) of e) is generated, A/D conversion is performed, and the corresponding average zero point data is subtracted from this A/D converted signal.0 When returning to the main routine, the measurement processing end flag is set to 1
", but now it is "0" and X weight is already being measured, so return to the beginning of the main routine, enter the interrupt routine as described above, and issue a measurement command. This is executed in determination 1 as to whether or not the edge of the pulse (pulse of the detector 6) is detected. Here, since edges are rarely detected, it is detected whether the measurement processing command flag is "1" or not.

Here, this flag was set to "1" last time, so A/
D 7'''-Yu is read, and the process is thereafter carried out in the same manner as the previous time.

When such processing is performed four times, the measured weight value integration register in the register 11 contains the value when the pulse generator 5φ generates four pulse signals after the detection pulse of the detector 6 rises. The A/D data minus the corresponding average zero point data is integrated. 4
When the first integration is completed, the measurement processing command flag is set to "0".
Set the measurement processing end flag to [month] and return to the main routine.

In the main routine, since the measurement processing completion flag is "1", the accumulated measured weight value is divided by 4 to obtain an average value, which is displayed on the display unit 14.

In the second embodiment, as shown in FIG. 5, the present invention is implemented without using a micromoxibustion computer. In the figure, 20-1 to 20-m are registers, which correspond to the integration registers of the first embodiment.

A counter 22 counts the pulses generated by the pulse generator, and corresponds to the timing counter of the first embodiment. 24 is also a counter, which counts the pulses generated by the rotation detector 7, and is similar to CH in the first embodiment.
This corresponds to the C counter. 26 is a temporary memory device,
Reference numeral 28 is an adder, 30 is a divider, 32 is a subtracter, and 3 is a weighted average arithmetic circuit, which corresponds to the arithmetic unit of the first embodiment.

Next, the operation will be explained. Now, registers 20-1 to 20-
m and the counter 24 have been reset, and the conveyor 1
rotates, and based on the pulses generated by the pulse generator 50, A
It is assumed that the /D converter 4 is outputting A/D data, and that the AND gate 35 is supplied with a dynamic zero point adjustment signal.

In this state, when the rotation detector 7 generates a pulse signal, the counter 22 is reset and the count value of the counter 24 becomes "1". At this time, the address pointer 3 is
6 designates register 20-1. The A/D data at that time is stored in the register 20-1 via the temporary storage register 26 and the adder 2. Next, when the pulse generator 5 generates a pulse, the count value of the counter 22 becomes "1".
, the address pointer 36 specifies the register 20-2, and the current gator is similarly stored. below,
Similarly, while the conveyor l rotates once, the register 20
-mK are sequentially stored.

When the belt conveyor 1 rotates once and the rotation detector 7 generates a rotation pulse again, the value of the counter 24 becomes 2, the counter 22 is reset, and the register 20-1 is designated again. As a result, the value stored in the 1 register 20-1 is called to the adder 28, and added to the value stored at that time in the temporary storage register 26, and the added value is returned to the register 20-1.
The score is returned to 0-1. Thereafter, integrated data is stored in the register 20-m in the same way. When this is repeated N times, the register 20-1 stores the N integrated values of the same section in one rotation of the belt conveyor. The same applies to other registers.

When the belt conveyor rotates N times, the dynamic zero point adjustment signal disappears, the AND gate 35 closes, and N becomes the divider 30.
is supplied from the counter 24. Then, according to the N+1 rotation, the integrated values are sequentially supplied from the registers 20-1 to 20-m, and the divider 30 sequentially divides them by NT and returns them to the respective registers 20-1 to 20-m. Note that the signal C supplied to the divider 30 is a division command signal, and is supplied when the count value of the counter 24 reaches N.

Next, when the belt conveyor l is transporting the article to be weighed, a measurement command signal is supplied to the AND gate 38, and the AND gate 38 is opened. When the one-rotation detector 7 generates a rotation pulse in this state, the counter 22 sequentially calls the registers 20-1 to 20-m. When the article to be weighed reaches the weighing section 2 and the detector 6 generates a detection signal, the subtracter 32 subtracts the stored value of the register called up at that time from the data output from the transducer 4, and weighs the weight. It is supplied to the plain cloth calculation circuit 4. The subtracter 32 performs this subtraction every time the pulse generator 5 generates a pulse signal while the detector 6 generates a detection signal. Weighted average calculation circuit 3
4 integrates and averages these subtracted values, outputs the average value, and displays it on the display section.

In the method according to the present invention, the belt conveyor is rotated once in an unloaded state, the total length of this one rotation is divided into a plurality of sections, the A/D conversion output of each section is stored, and the article to be weighed is conveyed. When carrying out the conveyance, the A/D variation force output of the section b is output without load A corresponding to this section.
Since the /D conversion output is subtracted, highly accurate measurement can be performed even if the dispersion of the distributed load on the belt is large or the eccentric load on the boob is large.

In both of the above embodiments, the belt conveyor is rotated N times in a no-load state to obtain the average zero point value for each section, but depending on the situation, the belt conveyor may be rotated only once. Still in the above embodiment one of the belt conveyors l
In cases where there is a possibility that the number of sections in rotation will not always be m due to the number of intervals and the transmission characteristics of gears, that is, the pulse generator 5 may generate m or more pulses during one rotation of the belt conveyor l. If there is a possibility, the number M is the maximum number of data that can cause the number of registers to be safe.
If the number of pulses from the pulse generator 5 exceeds m, the number of pulses from the pulse generator 5 exceeds m+1. A/D 7'-1 is stored in the m+2.......th register, and when the number of pulses of the pulse generator 5 is m, the remaining registers are all stored with the A/D 7'-th.
/D data (in the second embodiment, A/D data of the temporary storage register 26) may be written. Further, although four differences between the A/D data and the average zero point data are averaged and displayed on the display section, the number may be changed depending on the situation. Furthermore, although the pulse generator 5 and rotation detector 7 are provided separately, the pulses and rotation detection signal may be generated by one device. Further, in the above embodiments, the present invention was implemented on a belt conveyor, but it can also be implemented on a chain conveyor. Further, in the above embodiment, the weighing section 2 is provided at the lower part of the belt conveyor I, but the entire belt conveyor including the driving section may be weighed by a weighing machine.

[Brief explanation of drawings]

FIG. 1 is a block diagram of the device used in the first embodiment of the dynamic weighing method according to the present invention, FIG. 2 is an output waveform diagram of each part of the device in FIG. FIG. 4 is a flowchart of the main routine of the example, FIG. 4 is a flowchart of the interrupt routine of the first embodiment, and FIG. 5 is a block diagram of the device used in the second embodiment. l...Belt conveyor, 3...Load cell, 4...
- A/D converter, 5... pulse generator. Patent applicant Yamato Seiko Co., Ltd. Agent Tetsu Shimizu and 2 others (d)
]-95- ￥3 figure

Claims (1)

[Claims] (1) The Kl amount conveyor is rotated at least once without any load being applied, and the total length of the weight conveyor in this one rotation is divided into a plurality of sections, and the no-load-applied weighing signal for each section is memorized. and a process of calculating the difference between the article weighing signal in the section where the article is being conveyed and the no-load-applied weighing signal corresponding to that section in the article conveyance state of the weighing conveyor. Weighing method.