Failure detection device for spindle head of machine tool including a plurality of sensors

Patent metadata

Field	Value
Publication number	US-10364840-B2
Application number	US-201815959952-A
Country	US
Kind code	B2
Filing date	Apr 23, 2018
Priority date	Apr 28, 2017
Publication date	Jul 30, 2019
Grant date	Jul 30, 2019

Abstract

A failure detection device includes a formaldehyde sensor that detects the generation of formaldehyde, a vibration sensor that detects vibration of a spindle head, and a machine controller that detects the occurrence of an abnormality of a spindle bearing. The machine controller determines the occurrence of the abnormality of the spindle bearing and a cause of the abnormality based on a change in the output value of the formaldehyde sensor over time and a change in the output value of the vibration sensor over time.

First claim

The invention claimed is:

1. A failure detection device for detecting an abnormality of a spindle head of a machine tool, comprising:
a first sensor that detects generation of a predetermined first gas in the spindle head;
a second sensor that detects a change in a state of the spindle head that is different from the generation of the first gas; and
a controller that determines whether a bearing arranged in the spindle head is abnormal or not, wherein
the controller is configured to determine occurrence of the abnormality in the bearing and a cause of the abnormality based on a change in an output value of the first sensor over time and a change in an output value of the second sensor over time,
the bearing is a spindle bearing that supports a spindle,
the spindle bearing has a cage made of phenol resin and a lubricant disposed in the spindle bearing,
the first sensor is a gas sensor or an odor sensor that detects formaldehyde,
the second sensor is one of a vibration sensor that detects vibration of the spindle head or a sensor that detects an elastic wave generated in the spindle head, and
the controller determines that the spindle bearing is abnormal due to deterioration of the lubricant if the output value of the second sensor increases after the output value of the first sensor increases.

2. The failure detection device according to claim 1 , wherein the controller includes a display unit that displays a member in which the abnormality occurs and the cause of the abnormality.

3. The failure detection device according to claim 1 , wherein
the controller includes a machine learning unit that performs supervised learning,
the machine learning unit including:
a state observation unit that observes a state variable including at least one of the output values of the first sensor and the second sensor, increase rates of the output values, increase amounts of the output values, and a time from a start time of an increase in the output value of the first sensor to a start time of an increase in the output value of the second sensor;
a learning unit that acquires teacher data including information on occurrence of the abnormality and updates a learning model, based on the teacher data, for determining whether a predetermined member is abnormal or not; and
a decision unit that acquires a current state variable and determines whether or not the abnormality occurs based on the current state variable and the learning model.

4. The failure detection device according to claim 1 , wherein
the controller includes a machine learning unit that performs supervised learning,
the machine learning unit including:
a state observation unit that observes a state variable including output values of a plurality of sensors and at least one of data relating to an operation state of a spindle motor or data relating to a state around the spindle head;
a learning unit that acquires teacher data including information on timing of occurrence of the abnormality and updates a learning model, based on the teacher data, for predicting timing when a predetermined member becomes abnormal; and
a decision unit that acquires a current state variable and predicts the timing of occurrence of the abnormality based on the current state variable and the learning model.

5. A failure detection device for detecting an abnormality of a spindle head of a machine tool, comprising:
a first sensor that detects generation of a predetermined first gas in the spindle head;
a second sensor that detects a change in a state of the spindle head that is different from the generation of the first gas; and
a controller that determines whether a bearing arranged in the spindle head is abnormal or not, wherein
the controller is configured to determine occurrence of the abnormality in the bearing and a cause of the abnormality based on a change in an output value of the first sensor over time and a change in an output value of the second sensor over time,
the bearing is a motor bearing that supports a shaft of a spindle motor,
the motor bearing includes urea grease disposed in the motor bearing,
the first sensor is a gas sensor or an odor sensor that detects nitrogen oxide,
the second sensor is one of a vibration sensor that detects vibration of the spindle head or a sensor that detects an elastic wave generated in the spindle head, and
the controller determines that the motor bearing is abnormal due to deterioration of the urea grease if the output value of the second sensor increases after the output value of the first sensor increases.

6. A failure detection device for detecting an abnormality of a spindle head of a machine tool, comprising:
a first sensor that detects generation of a predetermined first gas in the spindle head;
a second sensor that detects a change in a state of the spindle head that is different from the generation of the first gas; and
a controller that determines whether an oil seal arranged in the spindle head is abnormal or not, wherein
the controller is configured to determine occurrence of the abnormality in the oil seal and a cause of the abnormality based on a change in an output value of the first sensor over time and a change in an output value of the second sensor over time,
the oil seal is disposed so as to seal an inner space of a spindle motor and includes an elastic member that is in contact with a shaft of the spindle motor and is made of nitrile rubber,
the first sensor is a first gas sensor or a first odor sensor that detects hydrogen sulfide,
the second sensor is a humidity sensor that detects humidity in the spindle head, and
the controller determines that the oil seal is abnormal due to biting of a foreign matter if the output value of the second sensor increases after the output value of the first sensor increases.

7. The failure detection device according to claim 6 , further comprising a third sensor that detects generation of nitrogen oxide in the spindle head, wherein
the controller is configured to determine occurrence of the abnormality in a bearing arranged in the spindle head and a cause of the abnormality,
the third sensor is a second gas sensor or a second odor sensor that detects nitrogen oxide,
the bearing is a motor bearing that supports the shaft of the spindle motor,
the motor bearing includes urea grease disposed in the motor bearing, and
the controller determines that the motor bearing is abnormal due to entry of liquid into the spindle head if the output value of the third sensor increases after the output value of the second sensor increases.

Assignees

• Fanuc Corp

Inventors

• Kamiya Yohei

Classifications

• F16C33/44

  Selection of substances (F16C33/40, F16C33/41 take precedence) · CPC title

• G05B19/406

  characterised by monitoring or safety (G05B19/19 takes precedence) · CPC title

• B23Q17/007

  for managing machine functions not concerning the tool · CPC title

• G05B2219/37435

  Vibration of machine · CPC title

• F16C19/525

  related to temperature and heat, e.g. insulation · CPC title