Vibration Sensing
A wireless mesh network of vibration sensors monitors factory machinery and provides early detection of failures.
The Client
Jose is an engineer at GL&V (now acquired by Valmet) and came to us with an idea to monitor machines in a paper processing facility.
The Challenge
A major issue in the paper machining industry is properly maintaining the massive multi-million dollar machinery that is used to churn out paper products 24 hours a day, seven days a week.
The objective of this project was to create a proof-of-concept to demonstrate the viability of monitoring machine health with a network of vibration sensors. We would use off-the-shelf parts like the nRF52840 for wireless mesh communication, the Particle Photon evaluation board for pushing IoT data to the cloud, and the ADIS16228 vibration sensor.
The Outcome
We created three devices to form the wireless mesh, with two devices being equipped with vibration sensors and mounted to paper machines and the third device acting as a gateway which would receive the sensor data and upload it to an online database where it could then be visualized with PowerBI software.
This was a successful proof-of-concept. We were able to upload and visualize the vibration data of factory equipment and compare the performance of the machinery between days or weeks. Ultimately, it was left up to GL&V to decide if the sensor data would be viable for monitoring machine health and in the end they did not decide to take the project further. However, it was definitely a rewarding experience for our company to bring many different technologies together and demonstrate the potential of IoT technology.
Technologies
Thread Protocol
Wireless mesh allows data to proliferate throughout the entire network of sensors.
MEMS Sensing
Tri-Axial MEMS sensors allows detailed analysis of vibration data to monitor machine health.
Microsoft Azure
Database management with flexible routing of information makes analyzing data possible.
Power BI
Front-end visualization software allows Azure data to be seen across multiple platforms.
Internet of Things
The Photon board by Particle was the perfect tool for getting a quick IoT solution for this proof-of-concept. Once it is connected to a WiFi router and given internet access, sending data to the cloud is simple.
For a real-world product, a custom solution would have to be developed due to the data fees associated with products like this one.
Vibration Sensing
The ADIS16228 takes twenty-thousand samples per second of vibrational forces (i.e. the acceleration) and it performs a fourier transform of the data so that its frequency spectrum can be analyzed.
Large industrial equipment like the machinery in a paper factory will usually start to exhibit irregular vibrational patterns leading up to a failure. These kinds of vibrations can be caused by damaged ball bearings, misaligned parts, or load imbalances.
Using a monitoring device like the one designed in this project can prevent a minor problem from going undetected and turning into a catastrophic failure.