JP3546953B2 - Inspection method for injection molding machine - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for determining the quality of a product of an injection molding machine.
[0002]
[Prior art]
In the injection molding machine, the quality of the obtained product is determined by various methods. As an example of a pass / fail determination method, a center value is calculated from actual value data for a plurality of shots in the past for a certain actual value obtained for each shot in a driving state in which a good product is obtained, and the center value is vertically determined with respect to this central value. Set a certain monitoring width. Then, for each shot, it is determined whether or not (center value-monitoring width) ≦ actual value ≦ (center value + monitoring width) is satisfied, and if satisfied, a good product is determined, and if not satisfied, a defective product is determined. ing. Note that the actual values include a weighing time, a cushion position, a pressure-holding completion position, a filling peak pressure, an injection peak speed, a pressure-holding peak speed, and the like. Normally, the center value and the monitoring width are set for all of these actual values, and the above determination is made for all of the actual values.
[0003]
[Problems to be solved by the invention]
However, in the above method, it is necessary to set the monitoring width by examining the variation of the actual value every time the product changes, and a large amount of work is required for the operator.
[0004]
Therefore, an object of the present invention is to provide a product quality determination method that can easily perform a setting operation for product quality determination in an injection molding machine.
[0005]
Another object of the present invention is to provide a method of determining the quality of a product, which can easily apply a set value determined for a certain product (a certain mold) to another product (a different mold).
[0006]
Still another object of the present invention is to provide a method of determining the quality of a product, which can improve the accuracy of determination.
[0007]
[Means for Solving the Problems]
In the method of determining the quality of a product of an injection molding machine according to the present invention, at least three of a plurality of types of actual values obtained for each shot of the injection molding machine are selected, and the past N shots are selected for each of the selected actual values. The center value is calculated from the actual value, and the upper and lower monitoring widths are set for the calculated center value. From the (N + 1) th shot, the monitoring width is updated for each shot, and the (N + 1) shot A first discriminating step of performing a discriminating operation to determine whether each actual value from the eye falls within a corresponding monitoring width, and discriminating a non-defective product if it is included; If it is determined in step that the candidate is a good item, from the (N + 1) th shot, the actual value of the immediately preceding shot and the actual value of the latest shot are subtracted from the at least three selected actual values. Subtraction result Te all positive or negative and if it defective candidates, and executes a second determination step of determining a non-defective otherwise.
[0008]
In this product quality judgment method, from the (N + 1) th shot, each of the at least three selected actual values obtained in each shot is averaged from the actual values of the past N shots counted from the immediately preceding shot. A value and a standard deviation are calculated for each shot, and a magnification ki is initially set with respect to the standard deviation calculated for the shot for the actual value of the shot determined as a defective candidate in the second determination step. It is determined whether or not (the actual value of the immediately preceding shot−the actual result value of the defective product candidate) ≦ (ki × standard deviation) is satisfied for each of the at least three selected actual values, and when all the actual values are satisfied, It is preferable to execute a third determination step of making a final determination of a non-defective product or a defective product.
[0009]
In this product quality judgment method, a moving average value and a standard deviation are calculated for each shot from the actual values of the past N shots counted from the immediately preceding shot as the center value in the first judgment step. Preferably, the monitoring width is updated and set for each shot by (moving average value ± K × standard deviation). In this case, the standard deviation in the third determining step is calculated in the first determining step. Standard deviation can be used.
[0010]
In this product quality judgment method, the actual values obtained for each shot further include a weighing time, a cushion position, a pressure holding completion position, a filling peak pressure, an injection peak speed, a pressure holding peak speed, and the like. , At least three actual values are selected, a moving average value and a standard deviation are calculated for each of the selected actual values, and a magnification ki with respect to the calculated standard deviation is also set for each of the selected actual values. It is characterized.
[0011]
Further, in the present product quality judgment method, the magnification is variable, and when a different product is molded and the product is similar in weight and shape to a previously molded product, in the third determination step, The optimal magnification ko obtained at the time of molding the previously molded product is set as an initial magnification ki.
[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described. The functions required to execute the product quality judgment method according to the present invention are as follows.
[0013]
Achievement value acquisition function: In order to determine the quality of a product, it is necessary to acquire various achievement values as in the conventional case. The actual values include the molding cycle time, the filling time, the filling pressure, the holding pressure speed, the V / P switching position, the holding time, the weighing time, the cushion position, the pressure holding completion position, and the injection peak speed as described above. There are pressure screw time (time during which the screw moves to a certain position during the pressure holding process) and the like, and these actual values can be acquired once per shot. For example, the weighing time can be known from the weighing timer, and the filling time can be known from the filling timer, and the cushion position, the pressure holding completion position, and the V / P switching position are measured by the screw position sensor. The filling pressure can be measured by a resin pressure sensor in the mold or a resin pressure sensor near the resin outlet in the heating cylinder, and the injection peak speed and the holding pressure speed can be measured by a speed sensor of the injection motor. In any case, such a function is also provided in a conventional injection molding machine.
[0014]
Function for storing and displaying actual values: A plurality of types of actual values obtained for each shot can be stored in a storage device provided in a control device of the injection molding machine, and provided in the injection molding machine as necessary. It can be displayed on a display or printed out by a printer. Of course, not only the actual value but also various pressure waveforms and speed waveforms for each shot can be displayed. A conventional injection molding machine has such a function.
[0015]
Standard deviation calculation function: This is a function unique to the present invention, and the average value and the standard deviation are calculated for each shot from the actual value data of the past N shots counted from the shot immediately before in a certain shot. That is, when calculating the average value and the standard deviation, the actual value data for N shots is not fixed, but the actual value data is updated by one shot for each shot. It is clear that the average obtained in this way is a moving average.
[0016]
Magnification setting function for standard deviation: This is also a function unique to the present invention, and is not a standard deviation itself calculated by the above calculation function but a value obtained by multiplying a standard deviation by a magnification k to perform good product quality judgment. I'm trying to use
[0017]
Pass / fail determination function: As will be described later, pass / fail determination is performed by first to third determination steps unique to the present invention.
[0018]
Next, the operation of the product quality judgment according to the present invention will be described. The procedure for making settings for product quality determination for a new product according to the present invention is as follows.
[0019]
First discriminating step In the first discriminating step, three (3) here are obtained from the above-mentioned plural types of actual values obtained for each shot of the injection molding machine after molding conditions are determined and a non-defective product is obtained. (Preferably three or more) is selected, a moving average value is calculated for each of the selected three result values from the result value data for the past N shots, and a standard deviation is calculated for each shot. Then, the upper and lower monitoring ranges for the three calculated moving average values are set as (moving average value ± K × standard deviation). From the (N + 1) -th shot, this monitoring width is updated for each shot. Subsequently, an operation of determining whether or not each actual value from the (N + 1) -th shot falls within the corresponding monitoring width is performed, and if it is included, it is determined as a good product candidate, and if not, it is determined as a defective product. N is a positive integer, preferably about 30, and K is preferably about 3. FIG. 3 shows an example of setting the monitoring width in the first determination step.
[0020]
In the first discrimination step, high discrimination accuracy is not required, so that (center value-monitoring width) ≦ actual value ≦ (central value + monitoring width) is satisfied for each shot, as in the conventional example described above. A determination may be made as to whether the product is a good product candidate if satisfied, or a defective product if not satisfied.
[0021]
Second discriminating step The second discriminating step is performed for each actual value of the shot determined as a good item candidate in the first discriminating step. That is, when it is determined in the first determination step that the result is a good candidate, subtraction is performed on the selected three actual values by subtracting the actual value of the previous shot from the actual value of the latest shot and the actual value of the latest shot. If all the results are positive or negative, it is determined to be a defective product candidate; otherwise, it is determined to be a good product. Of course, the subtraction in this step may be the subtraction of (the actual value of the previous shot)-(the actual value of the latest shot) or vice versa.
[0022]
In any case, a determination is made as to whether all the subtraction results for the three actual values are all positive or all negative, and if the same directionality of all positive or all negative is indicated, it is defective. Candidate, otherwise determined to be good. In other words, if at least one of the three actual values has a different direction from the other two actual values, for example, if one has a different direction, that is, one is positive and the other two are negative, it is determined to be good. .
[0023]
By the way, in the second discrimination step, even in the same direction, if the result of subtraction of all the actual values has the same direction in the + direction, the resin density is low, and there is a tendency for short shots. It has been confirmed that there is. Conversely, it has been confirmed that when the results of subtraction of all the actual values are negative, the resin has a high resin density and has a tendency to overpack, that is, to easily generate burrs, when the results have the same negative direction.
[0024]
FIG. 1 shows an example of a judgment material in consideration of directionality for three actual values of a measuring time, a pressure-holding completion position, and a filling pressure.
[0025]
Third determination step The third determination step is a final determination step, which is performed on the actual value of the shot determined as a defective candidate in the second determination step. For this reason, from the (N + 1) th shot, for the three selected actual values obtained in each shot, the average value and the standard deviation σ are calculated from the actual value data of the past N shots counted from the immediately preceding shot. Calculated for each shot. Here, as the average value and the standard deviation σ, the values calculated in the first determination step are used.
[0026]
Then, the magnification k with respect to the standard deviation σ is set as an initial magnification ki (where ki <K) based on past results of the same kind of product and statistical knowledge. Further, an initial magnification ki is set for the calculated standard deviation σ,
(Actual value of previous shot-Actual value of defective candidate candidate) ≦ (ki × standard deviation σ)
Is determined for each of the selected three performance values. If all of the performance values are satisfied, it is determined that the product is good, and if it is not satisfied, it is determined that the product is defective. (However, ko <K) is determined. Needless to say, this initial magnification ki differs for each actual value.
[0027]
Thereafter, the optimum magnification ko is set for the standard deviation σ calculated for each shot, and the actual value determined as a defective product candidate is
(Actual value of previous shot−Actual value of defective candidate candidate) ≦ (ko × standard deviation σ)
Is determined, and if all the actual values are satisfied, it is determined as a good product, and if not, it is determined as a defective product. As described above, in the third determination step, the optimum magnification ko with respect to the standard deviation is found while performing the molding / product quality determination.
[0028]
The products determined to be defective in the first and third determination steps are separated by a shooter of an injection molding machine or separated by a take-out machine.
[0029]
As described above, in the present embodiment, the quality of a product is determined through the first to third determination steps, thereby minimizing the possibility that a product determined to be a good product actually includes a defective product. It is also possible to reduce the possibility that non-defective products include non-defective products.
[0030]
FIG. 2 shows the result of a test performed on the presence or absence of a misjudgment in which a defective product is included in a product determined as a non-defective product, and it has been confirmed that there is no misjudgment even in a sample of several hundred shots.
[0031]
FIG. 4 shows a display example on a display provided in the injection molding machine for reference. Here, in addition to the weighing time, the cushion position, the pressure holding completion position, the injection peak speed, and the filling peak pressure, the minimum holding pressure speed and the filling pressures 1, 2, and 3 are used as the actual values. The filling pressures 1, 2, and 3 are values obtained by sampling the filling pressure at different timings in the filling step. Which of these performance values is used for the pass / fail judgment can be arbitrarily selected. When selecting, the monitoring column in FIG. 4 is set to “automatic”, and when not selected, “off” (“off” is shown in the drawing). )). On the display, an average value, a standard deviation, and a magnification k are displayed for each of the above-described actual values. Of course, the average value and the standard deviation may change for each shot. Further, the magnification k is different for each actual value. Furthermore, the actual value data for each shot is also displayed below these display portions.
[0032]
In the case where a different product is molded and the product is similar in weight and shape to the previously molded product, in the third determination step, the optimum magnification obtained in molding the previously molded product is determined. The default magnification ki can be set by diverting ko.
[0033]
【The invention's effect】
{Circle around (1)} When a product quality judgment is set for a new product, a magnification for the standard deviation is set. This is not a physical absolute value but a tendency determined by the product weight and shape (= dimensionless relative value), so that a general statistical idea can be used (for example, 99.7 within 3σ in a normal distribution). % Etc.). Further, when a different product is molded and the product is similar in weight and shape to the previously molded product, the optimal magnification ko obtained at the time of molding the previously molded product is set to the initial magnification ki and Can be set in many cases. That is, in many cases, the magnification k of a product having a similar product weight or shape can be used almost as it is.
[0034]
{Circle around (2)} The standard deviation is calculated using the actual values of the past N shots from the shot to be determined. This eliminates the need to perform a test in advance to determine the standard deviation.
[0035]
(3) As described above, the setting work for determining the quality of the product in the injection molding machine can be easily performed, and the magnification determined for one product (a certain mold) is diverted to another product (another mold). Can be made easier.
[0036]
{Circle around (4)} By performing the product quality determination through the first to third determination steps, it is possible to improve the determination accuracy.
[Brief description of the drawings]
FIG. 1 is a diagram showing an example of a judgment material in consideration of directionality for three actual values in a product quality judgment method according to the present invention.
FIG. 2 is a view showing a result of a test performed on the presence or absence of an erroneous determination in which a defective product is included in a product determined as a non-defective product in the product quality determining method according to the present invention.
FIG. 3 shows an example of setting a monitoring width in a first determination step in the product quality determination method according to the present invention.
FIG. 4 is a diagram showing a display example displayed on a display in product quality determination according to the present invention.

Claims (5)

At least three are selected from a plurality of types of actual values obtained for each shot of the injection molding machine, and for each of the selected actual values, a center value is calculated from the actual values of the past N shots. Monitoring widths above and below the center value are set, and from the (N + 1) -th shot, the monitoring width is updated for each shot, and each actual value from the (N + 1) -th shot falls within the corresponding monitoring width. A first discriminating step of performing a discriminating operation as to whether or not the discriminating part is present, and if not, discriminating it as a good product candidate;
If it is determined in the first determination step that the candidate is a good item candidate, from the (N + 1) th shot, the actual value of the immediately preceding shot and the actual value of the latest shot are compared with the at least three selected actual values. And a second determining step of determining a defective product if all the subtraction results are positive or negative, and a non-defective product otherwise. In the method according to claim 1,
From the (N + 1) th shot, for each of the at least three selected actual values obtained in each shot, the average value and standard deviation from the actual values for the past N shots counted from the immediately preceding shot are calculated for each shot. Calculate,
For the actual value of the shot determined as a defective candidate in the second determination step, the magnification ki is initially set with respect to the standard deviation calculated for the shot (the actual value of the immediately preceding shot−the defective candidate). A determination as to whether or not (actual value) ≦ (ki × standard deviation) is satisfied is made for each of the at least three selected actual values. A method for judging the quality of a product of an injection molding machine, wherein a third judging step of judging is performed. In the method according to claim 2,
As the center value in the first determination step, a moving average value is calculated from the actual values of the past N shots counted from the immediately preceding shot, and a standard deviation is calculated for each shot. (Value ± K × standard deviation), update setting for each shot,
A method for determining the quality of a product of an injection molding machine, wherein the standard deviation in the third determining step uses the standard deviation calculated in the first determining step. In the product quality determination method according to any one of claims 1 to 3, the actual values obtained in each shot include a weighing time, a cushion position, a pressure holding completion position, a filling peak pressure, an injection peak speed, and a pressure holding peak speed. At least three performance values are selected from these performance values, a moving average value and a standard deviation are calculated for each of the selected performance values, and a magnification ki with respect to the calculated standard deviation is also selected. A method for determining the quality of a product of an injection molding machine, wherein the method is set for the actual value of the injection molding machine. The product quality determination method according to any one of claims 2 to 4, wherein the magnification is variable, in the case of molding a different product, and in a case where the product is similar in weight and shape to a previously molded product, In the third judging step, a product quality judgment method for an injection molding machine, wherein an optimum magnification ko obtained at the time of molding the previously molded product is set as an initial magnification ki.

Images