Data processing: measuring – calibrating – or testing – Measurement system – Performance or efficiency evaluation
Reexamination Certificate
2002-05-13
2004-08-17
Nghiem, Michael (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system
Performance or efficiency evaluation
Reexamination Certificate
active
06778942
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
Conventionally, machine accuracy must be enhanced to increase yield of products manufactured by a machine tool. In order to monitor conditions while the machine is operating, an abnormality is detected either by directly measuring mechanical displacement of a working drive shaft or by measuring a drive current. The present invention relates to a monitoring method for a machine tool, in which the drive current is measured in a continual fashion so as to learn the condition of the tool.
The present invention is applied in a tool performing a cyclical operation of operating, standing by, operating, standing by. Examples of such tools include an injection mold, a cutting device, a press, a conveyor device and the like.
2. Description of the Related Art
Conventionally, when an object is to be machined by a machine tool, a copying operation is performed based on design dimensions. When this is performed, the value of the drive current is measured and monitoring is performed to monitor whether the drive current enters an abnormal range. For example, in a case where a cylindrically shaped manufactured product shown in
FIG. 12
is to be produced, the rotation of a drive motor is started by a machining start signal to make a machining table start moving.
The starting current of the machine tool drive motor is set at a high value so that the drive motor is maintained in a no-load state and the current flows at a fixed level until the areas of the product which are to be machined are set.
As shown in
FIG. 11
, the product has areas which are machined by different tools A, B and C, and the frequency with which each tool is switched varies according to the wear conditions of each tool. Further, the switching of the tools is performed in the state when there is no machining load.
When the machining ends, the motor rotation is topped by also using a break. Therefore, the value of the electric current increases.
The machining conditions are managed as follows. When a drive motor having an output which is appropriate for the product is used, machining operation affects a current supplied to the drive motor. Thus, the machining conditions can be monitored by monitoring the current to the drive motor. Accordingly, it is possible to obtain data to raise the precision level of the product or to improve yield.
An example method for monitoring 1 cycle (all sections of 1 cycle) of the machining of the product is as follows. Equally spaced graduations are made along a time axis and measurement points are set at these graduations. Measurement values are stored as a time series. Digital data from a first cycle is used as a temporary standard of reference to be compared against the digital data from the next repeated cycle with regard to each sampling point, and a greatest value and a smallest value regarding each sampling point is stored. Thereafter, measurements are performed repeatedly, and operations are performed to replace the greatest values and the smallest values as needed to obtain the greatest value and the smallest value from each sampling point. Then, as shown in
FIG. 12
, data of the greatest values and data of the smallest values for all points are strung together and a pattern of a greatest value waveform (upper limit) M and of a smallest value waveform (lower limit) m are set, whereby machining abnormalities can be monitored visually.
That is, the greatest value waveform M and the smallest value waveform m are displayed on a display section of the machine device, and when the work operation is being performed, the actually measured values are visually checked within this pattern. Thus, an abnormality can be detected when an actual measured waveform R crosses the greatest value waveform M or crosses the smallest value waveform m.
Thus, data from for example 1000 cycles are drafted out one on top of the other to obtain the greatest value waveform and the smallest value waveform, whereby the monitoring of the machining status becomes effective.
However, when machine parts are machined in an automobile factory, machining operations easily exceed 1000 cycles in one day. That is, when the time used to machine one part is 60 seconds, the number of cycles in 24 hours of continuous operation is 60×24=1440 cycles. With this many machining cycles, 1000 sample numbers is insufficient to learn the state and precision level of the machine. This is because when the number of sample measurements is too small, elements of chance become great in the measurement operation, such as an occurrence of a random large (or small) value, or occurrence of atypical stability during the measurement. Therefore, it would be desirable to take an extraordinarily large number of sample measurements; however, when using the method of drafting out the data on top of each other, there is an opposite effect such that when an extraordinarily large number of sample measurements is taken, data which is taken during times of instability or taken under bad conditions is mixed in as well, thus making it difficult to perform precise monitoring.
Further, in the conventional art, the permitted range for the machining was set as a succession of upper limit data and a succession of lower limit data. As a result, a quick response could not be made when the abnormality was detected. When monitoring was attempted by taking many sampling points from complex machining areas of the product occurring in one cycle, data from other complicated machining areas of the product and data which takes the entire product into account become insufficient. Therefore, it has been necessary to provide more monitoring devices which are set with multiple settings.
SUMMARY OF THE INVENTION
The present invention has as an object to provide a monitoring method for a machine tool, in which sample measurement points are arranged in an appropriate manner for each complex machining area and for each simple machining area of a product machined in a work operation process to thereby prepare monitoring data and in which work machining is monitored using the monitoring data.
In order to achieve the above-mentioned object, the invention according to a first aspect of the present invention is characterized in that: an amount of change in one cycle from the beginning until the end of the operation process is converted into a readable signal; sampling pointss are set in response to changes in the signal, and sampling data that are measured across a plurality of cycles are saved; and for each sampling point a standard deviation value is obtained and program processing is performed; and the actually measured values obtained on the signal are compared against the standard deviation values to monitor the presence/absence of an abnormality in the operation process.
In one example of the above-mentioned configuration, at important parts of one cycle of the signal change the signal change is sampled at short intervals (ex., 1 ms), and at parts of the cycle which are not important the signal change is sampled at longer intervals (ex., 10 ms), and these data are saved.
According to a second aspect of the invention, in the invention according to the first aspect, the actually measured values of the work operation are monitored at each sampling point, and depending on whether the actually measured values are in a normal or an abnormal state, program processing is performed using the data from the sampling point.
According to a third aspect of the invention, there is provided a method for monitoring a load current supplied to a machining motor of a work machining device in 1-cycle units running from the start of the machining to the end of the machining, characterized in that for areas where the shape of the work to be machined is complex, sampling points are set at shorter intervals along a time axis of the machining, and for simple machining areas the sampling points are set at longer intervals along the time axis of the machining, and the sampling data at each sampling point are individually stored into a CPU stor
Nghiem Michael
Softrox Co., Ltd.
LandOfFree
Monitoring method for machine tool does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Monitoring method for machine tool, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Monitoring method for machine tool will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3316105