FEMM Hub Logo

X2: In-process tracking and tractability for zero-defect manufacture of electrical machines

In-process tracking and tractability for zero-defect manufacture of electrical machines

This project is centred on the numerous manual processes that underpin electrical machine manufacture, and how non-destructive testing and manufacturing digitisation methods can be integrated to provide added value to the manufacturing life-cycle. Through such technologies and methods, our aim is a zero-defect manufacturing approach in the production environment for electrical machines where we can move away from end-of-line test and towards in-process inspection, verification and digital certification of parts and processes.


The manufacture of electrical machines is a synthesis of traditional machining/forming and cutting processes interspersed with assembly, integration and test. In low and medium volume machine manufacturing, particularly for machines with the very highest levels of performance, input from skilled manual processes is commonplace. 

In-service failure arising from defects in electrical machine manufacture can be driven by one or more of these manual processes, manufacturing tolerances and variability in the constituent production processes. Arguably the most challenging aspects here are the fitting of coils into stator core, the connection of coils to the machine terminals and the various wet-processes involved in coil manufacture such as varnishing or encapsulation. 

The vast majority of quality assurance during manufacture is through inspection or test at various steps during manufacture and assembly. Such established approaches do not tend to capture or make use of in-process data for tracking and traceability of parts towards defect detection, mitigation and certification at the point of manufacture.

In-service failure of electrical machines can often be catastrophic, leading to total loss of the machine as a result of electrical shorts through insulation or terminal contact failures. Sub-optimal temperature profiles within the machine during operation and degraded performance of the electromagnetic characteristics can also result from manufacturing process variation. 

Through better tracking and traceability of the processes, we can mitigate these failures by identifying their causes earlier, rectifying them and removing the need for expensive end-of-line tests and certifications.

Current progress

The research team is currently specifying the scope for a sensorised workbench that can enable in-process tracking and tractability for manual processes in electrical machine manufacture. A comprehensive review paper is also being drafted in this area.

Future plans

Development of a sensorised workbench: The premise of this task is to capture and digitise key process characteristics that lead to downstream defects. Our plan is to study inspection procedures to create a full map of possible defects and the process characteristics during manufacture that lead to these defects. 

This will be followed by a comprehensive failure mode and effects analysis (FMEA) to detect the onset of failures and identify the possible digitisation system necessary to intercept the developing defects and the causal factors. The goal is to develop a ‘sensorised workbench’ to capture process characteristics by using networked sensors employing multiple sensing modalities (vision, pressure, temperature, noise etc) to simultaneously track human and process data.

Staff involved

The team involved with developing this goal come from varied research backgrounds, with knowledge and experience in digital manufacturing, manufacturing processes, sensor development, networks and communications, optimisation and human-workpiece interactions. This includes Professor Ashutosh Tiwari, the Airbus/RAEng Research Chair in Digitisation for Manufacturing, and Dr Michael Farnsworth and Dr Divya Tiwari, research associates in the field of digital manufacturing. Our current team also includes PhD student Ze Zhang.

If you would like more information on this project, contact Dr Michael Farnsworth.