Remote Machine Monitoring

Abstract

Clippard, A well-known company in Cincinnati Ohio that manufactures pneumatic valves, cylinders, and fittings, provided us with a challenge to build a system for them. This system will monitor 16 CNC machines during the off hours as they only work 1st shift. They are relying on the machines to stay running thought-out the afternoon and night. When a CNC machine goes down it stays down till the morning when a technician comes in to restart it. This leaves a substantial amount of unnecessary down time for the company. They have requested a system to be designed and built to alert a technician by text or email when either a machine is down or if the fire suppression system is triggered. Our group brainstormed several different ideas of how to get two signals to a computer through the internet to phone or e-mail. Given the company’s environment around each machine, we concluded that wireless signals through the WiFi system would be the best fit. The machine monitoring system will work with an ESP8266 NodeMCU that will be wired into each machine to communicate with the server. Using a digital input on the chip (GPIO pins) will give us our signal. The ESP8266 chip is programmable to connect to Wi-Fi then using IEE 802.11 protocol. This signal will be a publish only to the MQTT Mosquitto broker which is installed on the Raspberry pi. Node-Red will then receive and translate the signals to be displayed on the dashboard and be sent as notifications in the form of email or text message. Node-Red is a flow-based development tool for visual programming. Node-Red will process the data and if an alert is detected, it will send the email or text as it is programmed to do. The Mosquitto MQTT broker along with Node-Red will be hosted on a Raspberry Pi 4. The Raspberry Pi a low cost, credit-card sized computer that plugs into a computer monitor or TV and uses a standard keyboard and mouse.