Top 3 Motor Failure Detection Sensors & Benefits - How To Predict Machine Failure

Top 3 Motor Failure Detection Sensors & Benefits

How To Predict Machine Failure​

It’s no secret that motors run our lives. But some motors are far more important than others, and in some cases, motor failure is not an option.  Keep in mind, we rely on motors to run the cooling systems in nuclear reactors.  We rely on motors to run elevators in high-rise buildings.  Motors keep fuel flowing into jet engines of our airplanes.  There’s a lot riding on today’s motors, and failure can cost lives.  Because of our reliance on motors, motor failure detection is probably more important today than at any point in our history.

In these kinds of applications, it would be an engineering catastrophe to look at motors simply as output devices, as this akin to ignoring the other 50% of their capabilities.  The fact is, motor are on the receiving end of numerous counter-forces.  Motors should be looked upon as the primary sensor element in any predictive maintenance strategy.  Motors can tell you a lot about the health of a machine, so taking the time to listen to what motors are saying only makes perfect sense. 

NCD CEO Ryan Sheldon and CTO Anil Bhaskar teamed up with TVT America‘s CEO Mark Solvie to take a deep dive into the characteristics associated with motor failure.  NCD is now actively working with TVT to put together a sensor package designed specifically for motor failure detection and prevention in common 3-phase industrial motors.  Here’s a look at the 3 most important sensors we are focused on and why they they are so important in motor failure detection.

1. Vibration Sensors

Vibration sensors are at the top of our list for motor failure detection.  We are actively working on ways to listen to micro-vibrations in motor movement to determine the characteristics of motors under normal operation.  Keep in mind, when motors start and stop, these kinds of vibrations are often violent, but still within normal operation.  Detecting unusual vibrations that are not characteristic of usual operation often requires the assistance of cloud-based computers.  Our objective is to provide the raw data so that unusual vibrations can be detected.  Getting the right balance of meaningful data and low-power is important.  Our V2 Mems vibration sensor is fine tuned to handle a wide range of vibration applications with a long battery life and the ability to deliver time series data. Using this sensor to monitor unusual motor vibration can indicate bearing failure, load imbalances, motor stalls, and motor free-wheeling.  Wireless vibration sensors are a critical component to any motor health monitoring strategy.

2. Thermocouple Sensors

While most industrial motors include an integrated temperature sensor, we like to use our own temperature measurements to control temperature monitor accuracy.  Temperature measurement can be a great indicator of motor loads.  It’s no secret that motors may run hot or cool, and comparing these data to ambient air temperature can help predict normalized temperature trends.  Keeping an eye on motor temperature is obviously a good idea, but looking for changes in temperature trends over the lifespan of a motor is probably the most beneficial in plotting a motor failure detection strategy.  It’s not enough to say a motor is hot or cold, what’s important is keeping track of temperature patterns and looking for unexpected trends.  Overlapping these data with vibration data can bring a potential motor failure event into focus.  We suggest pairing these sensors with a great AI platform such as Microsoft Azure to help plot trends.

3. Current Sensors

In an effort to obtain a more complete picture of how motors are operating, monitoring the current of at least one leg is always a good idea.  It’s obviously a good idea to watch for current spikes to determine if a motor is under undue stress.  Similarly, current sensors can also be a great indicator of a motor that is underperforming due to free-wheeling.  This could indicate a broken belt or chain in the drive system.  Once again, overlapping the vibration, temperature, and current consumption data can provide a relatively complete picture of normal and abnormal motor operation. 

You may notice that all three of these sensors have one common purpose: they keep their focus on the motor directly. The benefits of focusing on the motor itself cannot be understated.

By combining all three of the above sensors together into a single device, we created a predictive maintenance sensor to help users cut costs.  Our predictive maintenance sensor combines a vibration sensor (which also includes ambient air temperature sensing), with a 100 Amp current monitoring sensor, and a Thermocouple sensor.  Combining all sensors into a single device means more data points are communicated in a much smaller installation profile.  Predictive maintenance IoT sensors also offer an equally impressive battery life of up to 500,000 wireless transmissions, which may be spread over 10 years.  Predictive maintenance sensors send data on user-preset intervals, allowing users to configure the battery life according to installation needs.  We are always looking for ways to help our customers cut costs, and by combining our most popular sensors into a single package makes a lot of economic sense.  With a 2-mile range and the ability to communicate directly to Amazon AWS, Microsoft Azure, Losant, Temboo, and MQTT Servers, this device is ready to communicate to the cloud.  For those who need localized data, we also support USB, Ethernet, RS-485, Bluetooth, RS-232, and WiFi modems.