Pick and place machines(PNP) are robotic assemblies that use a vacuum to lift a component off of a piece of tape, rotate it to the correct orientation, then place it on a circuit board. Because they run at really high speeds and must be able to put pieces with sub-millimetric precision, any imbalance of the belt-pulley system used to drive the working head or any vacuum pump malfunction directly translates into components being misplaced and completely ruining the workpiece.
We will be developing a predictive maintenance solution that gathers and analyzes vibration data, mounted on the carriage of a PnP machine, that can signal an operator when an anomalous behavior in the nominal functioning of the machinery is detected.
Predictive maintenance offers an alternative to usual maintenance procedures, helping to minimize the presence of specialized teams on the factory floor, reduce the inventory costs, and alerting the plant manager when machinery starts functioning outside nominal parameters before it endangers both the quality of the job in progress as well as the integrity of the machinery.
Built for Edge Impulse by the Zalmotek team