JPH0295215A - Counting scale - Google Patents

Abstract

PURPOSE:To recognize the circumstances of the occurrence of errors in counting at the first glance by providing two display units and allowing one of them to display the number of samples obtained by dividing the total weight of the sample on a dish by simplex value and the other to display the number of them obtained without cumulating fractions in counting which emerge every time the sample is added respectively. CONSTITUTION:The weight of sample M is obtained based on the data from a load sensor (a) by a sample weight calculation means (b). The number of sample K is calculated by dividing the weight of sample M by the simplex value mu previously stored in a simplex memory (c), which is rounded by a number calculation means (d). The calculated number of sample K is displayed on the 1st display unit (e). The displayed contents of the 2nd display unit (j) and the weight of sample M at the point of time when a command is issued from a command issuing means (f) are stored in a 1st memory (g) and a 2nd memory (h) respectively. In an operation means (i), the value obtained by subtracting the contents M0 of the memory (h) from the weight of sample M at present is divided by the contents mu of the memory (c) and is rounded. Then, the contents K1 of the memory (g) is added to the rounded value and the operated result is displayed on the display unit (j).

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention calculates and displays the number of samples by dividing the sample weight based on the output of a load sensor by the weight per sample (unit weight value). It is related to so-called counting scales.

<Prior Art> In the field of counting scales, many proposals have been made for the purpose of improving counting accuracy. Most of these are mainly related to improving algorithms to eliminate weighing errors of the scale itself and to more accurately determine unit weight values.

By the way, if the weight dispersion of the samples to be counted is greater than a certain level, the occurrence of counting errors cannot be prevented even with the conventional proposal described above due to factors such as accumulation of rounded numbers when calculating the number of samples. I can't.

Therefore, in order to avoid such counting errors, conventionally, the number of samples sampled to obtain the unit weight value is doubled.
Place about 2.5 times as many samples on the plate and note the number, and each time take the sample down and transfer it to another container, etc., and then place the next sample of about the same amount again. Measures are taken such as repeating the process and adding up the number of notes taken at the end.

<Problem to be solved by the invention> The conventional measures described above not only complicate the work, but also require 2 to 2.5 times the number of pieces, based on intuition.
There is no guarantee that counting errors will not occur.

In addition, as a proposal aimed at obtaining better counting accuracy in cases where only inaccurate unit weight values are obtained due to large variations in sample weight, etc., for example, Japanese Patent Application Laid-Open No. 58-204326 There is a method of statistically calculating how much counting error is included in the counting results, but the calculation is complicated and the software of the microcomputer built into the counting scale is complicated. There is a problem that the burden becomes heavy.

The present invention was made in view of these points, and it is possible to calculate the occurrence of counting errors as a guideline even if a unit weight value with poor accuracy is used, without burdening the microcomputer with complicated calculations. The purpose of the present invention is to provide a counting scale that allows you to know at a glance, and allows you to easily carry out appropriate counting operations with the required precision without having to perform complicated operations.

Means for Solving the Problems> The configuration for achieving the above objects will be explained with reference to the basic conceptual diagram shown in FIG. A sample weight calculation means for calculating M, and a number calculation means for calculating the number of samples K by dividing the obtained sample weight M by the unit weight value μ stored in advance in the unit weight memory C and rounding the result. d, the scale has a first display e that displays the calculated number of samples K, a command generation means f, and a first display that stores the display contents of the second display j below at the time of command generation. A memory g, a second memory h that also stores the sample weight M at the time when the command is issued, and a value obtained by subtracting the content M0 of the second memory h from the sample weight M at the present time, and the content μ of the unit weight memory C. By providing an arithmetic means i that divides the result by dividing the value by , and adds the rounded result to the contents of the first memory g, and a second display j that displays the result of the arithmetic operation,
characterized.

Note that the command generating means f may be a switch for the measurer to generate a command at any time, or a means for automatically generating a command upon detecting that the output of the load sensor a is in a stable state. be.

<Function> The first display e shows the number of samples K obtained by dividing the total sample weight M loaded on the load sensor a by the unit weight value μ, that is, = (M /μ rounded value) is displayed.

On the other hand, the second display j shows the number of pieces obtained by dividing the additional weight (M Mo) from the time the command was issued by the unit weight value μ, that is, the number of additional pieces and the number of loaded pieces at the time the command was issued. + is displayed in the total number ' = ((M-Mll)/μ rounded value) +. In other words, the value displayed on this second display j is
is the sum of the number already rounded before the command was issued, and the number calculated based on the weight added after the command was issued, and the fractional parts that occur in the counting operations before and after the command was issued are accumulated. It can be regarded as the number obtained without

It is not necessarily clear which one represents the correct number, the value displayed on the second display j, or the value displayed on the first display e, where the fractional parts are accumulated, but The difference between the displayed values and the displayed values can be a measure of the degree of counting error included in either value. In other words, for example, when the difference is 0, the counting error is 0, and when the difference is 1, the counting error is within ±1. You can know the number of items that can be counted at one time.

<Example> FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention.

The load detection section 1 includes a load sensor that engages with the plate 1a, an AD converter for digitally converting the output of the sensor, and the like.

Load data from the load detection section 1 is input into the microcomputer 2 at predetermined minute intervals.

The microcomputer 2 includes a CPU 21, a ROM 22,
It is composed of a RAM 23, an input/output port 24, etc., and a program to be described later is written in the ROM 22. In addition to a work area, the RAM 23 also stores the tare amount T, unit weight value μ, and information at the time of command generation, which will be described later. In the calculated number of pieces, and the sample weight value M. Areas are set up to store each item, etc.

In addition to the load detection section 1, the input/output port 24 of the microcomputer 2 includes first and second displays 3 and 4 that digitally display counting results, a tare switch 5, and a unit weight value setting switch 6. A command switch 7 for issuing commands, which will be described later, is also connected.

The tare switch 5 and the unit weight value setting switch 6 are well-known ones, and when the tare switch 5 is pressed, the weight on the plate 1a at that time is stored in the RAM 2 as the tare amount T.
This switch is used to set the sample weight M to be the weight obtained by subtracting the stored content T from the weight on the plate 1a.The unit weight value setting switch 6 is used to set a predetermined number of samples, for example, to the sample weight M. By placing the sample on the plate 1a and pressing it, the unit weight value μ is calculated by dividing the sample weight by the predetermined number and stored in RAM2.
This is a switch for storing data in 3.

The command switch 7 is a switch for, when pressed, storing the sample weight M on the dish 1a at that point in time in the RAM 23 as Mo, and at the same time storing the displayed value on the second display 4 at that point in time as KI. It is.

In addition, the first and second displays 3 and 4 each display the number of samples on the dish 1a calculated by the microcomputer 2, but the calculation method of the number of samples displayed on each display 3 and 4 is are different from each other as shown below.

Figure 3 shows the programs written in the ROM 22.
This is a flowchart showing the contents of the counting routine, and the operation will be described below with reference to this figure.

Note that this counting routine is performed using the unit weight value switch 6 described above.
It starts after executing the unit weight value setting routine that calculates and stores the unit weight value μ by the operation.

When this counting routine starts, first, the tare amount T in the RAM 23 and the above-mentioned Ni 1 and Mo are set to 0 (STI), and then loading of load data is started (
Sr2). A plurality of areas for storing load data are set in the RAM 23, and the oldest data is discarded each time the latest data is taken in. Then, each time this load data is taken in, the total weight W loaded on the plate 1a is determined by averaging the load data in the RAM 23 (Sr1).

When the tare switch 5 is pressed, the total weight W at that time
is stored in the RAM 23 as the tare amount T, and thereafter, the weight obtained by subtracting the tare amount T from the total weight W is recognized as the weight M of the sample to be measured on the pan 1a. Note that when the tare switch 5 is pressed, the value of Mo in the RAM 23 is also set to 0 (Sr1).

Now, by dividing the sample weight M to be measured by the unit weight value μ in the RAM 23 and rounding off the fraction, the number K of samples on the dish 1a is calculated (Sr1).

Also, from the sample type i1M to be measured, the weight value M in the RAM 23
The value obtained by subtracting 0 (M Mo) is divided by the unit weight value μ, and the fraction is similarly rounded to a value of 2 (STIO).

Note that the weight value M0 is 0 while the command switch 7 is not pressed yet, and therefore, during this time, 2 and K are equal values.

Next, ' is calculated as the sum of K2 and ' in the RAM 23 (STII). Then, K is displayed on the first display 3, K'
is displayed on the second display 4 (ST12).

Here, as long as the command switch 7 is not pressed, it is set to O as well as M0 due to initialization, so during this period, 2 = 2'.

When the command switch 7 is pressed, the content displayed on the second display 4 at that time is '=Kl+Kt:K.

At the same time, the sample type IM on the dish 1a at that time is stored as Mo in the RAM 23 (Sr1, 5T9).

Therefore, after the command switch 7 is pressed, the number calculated at 5T10 will be 2, which will be 100% after the switch is operated.
It represents the sample weight (M Mo) added on a divided by the unit weight value μ and rounded, that is, the number of additional pieces calculated independently based on the added weight. At the same time, 5TII
The value calculated in and displayed on the second display 4 is the sum of this additional number of samples, 2, and the number of samples before the command switch 7 is operated.

That is, the second display 4 displays the cumulative total of each additional number obtained by dividing each additional weight value separated by each press of the command switch 7 by the unit weight value μ and rounding. Ru. Therefore, the displayed number represents the number calculated without accumulating the fractional parts that appear when calculating each additional number.

On the other hand, the first display 3 displays the number obtained by dividing all sample types NM on the dish 1a by the unit weight value μ and rounding the result, and represents the number obtained by accumulating the above-mentioned fractional parts. become.

The numbers displayed on the first and second displays 3 and 4 are equal to each other when the weights of the counted samples are almost the same centering on the unit weight value μ, but there is a variation in weight. If there is a possibility that a counting error will occur because of a large number of errors, a difference will occur depending on the degree of the error.

The measurer can therefore know the degree of counting error based on the magnitude of this difference.

A specific example will be given and explained below.

[Table] is an example of the individual weights of 30 samples created based on a random number table, and the average weight X is 1.571625 g
, the standard deviation σ is 0.044 g, and the coefficient of variation is 2.8%.

When counting such samples, it is quite conceivable that the unit weight value μ calculated by sampling n samples takes a value within the range of, for example, ±3σ15. When n-5, the unit weight μ falls within the range of 1.5126 g to 1.6307 g. Now, suppose that the unit weight value μ is found to be 1.513 g,
It is assumed that the samples are placed on the plate in order from 1lh1.

With conventional counting scales, the total weight of the sample on the plate is calculated as the unit weight value μ
Calculate the number by dividing by and rounding to the nearest whole number, so 11
The correct number is displayed until m1 to IIkl12 are counted sequentially, but when 11kL13 is added, it is displayed as 14. That is, (1,626+ 1.589+...+1.534)/
1.513=12.452: 12 pieces (1,626+1
．． 589+...+1.622)/1.513=13.
524=14 (becomes solid. From then on, numbers 14 to 11
Until h30, the numbers that are one more than the true number are displayed.

On the other hand, in the embodiment of the present invention, the following will occur.

For example, if the command switch 7 is kept OFF while the samples 11 to 10 are being placed on the sample 11, then the numbers displayed on the first display 3 and the second display 4 will match.

(1,626+ 1.589+...+1.516)/
1.513 = 10.42 = If you press the command switch 7 after placing 10 pieces of 1tio, the number of Mo
"" 15.774, K+=10 is stored. In this state, if you place N[Lll-Il & 120 and do not press the command switch 7 during that time, the number displayed on the first and second indicators 3.4 will be 11h13 between '. There will be a difference. That is, for example, when up to Nn20 is loaded, the sample weight M on the plate 1a is 1.626 +
Since 1.589+...+1.583=31.435 g, K t :(M Mo )/μ= (31,43
5-15,774)/1.513= 10.35=10
Therefore, K' becomes 20. On the one hand, there are 21
It is individual. This makes it possible to know that +1 counting error has occurred in the displayed number K or NI'.

Also, if you press the command switch 7 again after loading N1120, MO = 31.435, K, = 20 will be stored in the RAM 23, and if you then load l1h21-11h30, M = 47.149, M - MO =15.714
Therefore, K=10 is obtained, and K' becomes 30.

At this time, K is 31, and it can be seen that a counting error of ±1 occurs even in this state.

As described above, by repeatedly placing an appropriate number of samples on the plate 1a and pressing the command switch A, a difference between the two displayed values will occur. After adding samples and repeating the operation of command switch A several times, the number of samples on pan 1a when the difference between both displayed values is within the range of O is -
It can be seen that this is an allowable number that can be counted with a counting error of 0 by placing it at a certain degree. Furthermore, if a counting error of ±3 pieces is allowed, the number of samples on the dish 1a may be counted at once when the difference between both display values is within 2, for example, by repeating the same operation. I understand that.

As a result, the number of pieces to be placed on the plate 1a from now on is determined according to the allowable counting error.

In the above embodiment, K and ' are displayed on the first and second displays 3 and 4, respectively, but K is displayed on either one.
Alternatively, the difference between K and ′ may be displayed on the other side.

In addition, the command to store K+ and Mo in the RAM 23 may be generated artificially by operating the command switch 7, or may be issued, for example, five times in a row to indicate that the data from the load detection unit 1 is stable. It is also possible to make a determination based on a condition such as that the value falls within a specified fluctuation range, and to automatically generate the error based on this condition.

Further, if the configuration is configured such that this command is generated when the command switch 7 is pressed and the data is stable, it becomes possible to cancel fractions more precisely without errors.

<Effects of the Invention> As explained above, according to the present invention, two indicators are provided, and one displays the number obtained by dividing the total weight of the samples on the plate by the unit weight value, and the other displays the number of samples obtained by dividing the total weight of the samples on the plate by the unit weight value. is configured to display the number of pieces obtained without accumulating the counted fractions that appear each time a sample is added, and the situation where counting errors have occurred can be easily determined from the difference between these two displayed values. It is possible to know the number of objects that can be counted at one time according to the required counting accuracy without having to rely on intuition. Moreover,
Since there is no need to perform complex statistical calculations, no burden is placed on the built-in microcomputer.

[Brief explanation of the drawing]

Figure 1 is a basic conceptual diagram showing the configuration of the present invention, Figure 2 is a block diagram showing the configuration of an embodiment of the present invention, and Figure 3 is the ROM.
22 is a flowchart showing the contents of the program written in the computer. 1... Load detection section 2... Microcomputer 21... CPU 22... ROM 23... RAM 3... First indicator 4... Second indicator 7... Command switch patent applicant Agent of Shimadzu Corporation
Patent Attorney Nishi 1) Newly built drawing 1 drawing 2 drawing

Claims (1)

[Claims] A sample weight calculation means that calculates the sample weight based on data from the load sensor, and calculates the number of samples by dividing the calculated sample weight by the unit weight value stored in the unit weight memory in advance and rounding. The scale includes a first display for displaying the calculated number of samples, a command generation means, and a first display for storing the display contents of the second display at the time of generation of the command. a memory, a second memory that stores the sample weight at the time when the command is issued, and a value obtained by subtracting the content of the second memory from the sample weight at the current time by the content of the unit weight memory; A counting scale, comprising: arithmetic means for adding the contents of the first memory to a rounded value; and a second display for displaying the arithmetic result.