In CFRP bonded repair applications, it is mandatory to ensure surface cleanliness for achieving robustness and reliability of the bonds. Unfortunately, aircraft surfaces can be affected by several potential contamination sources that may hamper the bond strength; such contaminations have to be detected before a bonded repair takes place. Maintenance and repair operating conditions also require that any contamination detection technique should be operated on a portable basis and measurement results be readily available to the operator.

Artificial olfaction (AO) technologies could represent an optimal framework to develop such a detection tool (i.e. an electronic nose). They are based on the integration of gas sensing devices and pattern recognition (PR) algorithms. Their working principle mimics mammal’s olfaction systems: a gas sensor array, being exposed to a gas mixture, produces a distinct response pattern that can be interpreted by software components to produce qualitative and quantitative estimation on the mixture composition.

Our proposal addressed the investigation of AO technology suitability and the development of an e-nose prototype for surfaces contamination detection to be used in bonded repair of aircraft composite structures. The selection of an ad-hoc sensing array was carried out, investigating the use of several gas sensors based on different operating principles. Their capability to reliably react to volatiles emitted by potential aeronautics contaminants (e.g. skydrol, fuels, etc.) and to exhibit limited drift behaviour was firstly investigated in controlled environment. Once a response database was recorded, a PR component was designed and developed to provide qualitative and quantitative readings. Design and development efforts of a final prototype inspection tool was focused on portability in order to achieve operative capability.