Wrinkle characterisation from ultrasonic scans of composites

Beatriz Larrañaga-Valsero, Robert A. Smith, Rostand Tayong Boumda, Antonio Fernández-López, Alfredo Güemes

Research output: Contribution to conferenceConference Paper

265 Downloads (Pure)

Abstract

Wrinkles (out-of-plane waviness) are, unfortunately, a common phenomenon during composite manufacture and have a significant influence on the mechanical performance. While the aviation industry tries to reduce the thickness of composites for weight reasons, more plies are added to the design in order to mitigate the risk of undetected wrinkles and ensure robust mechanical properties of structures. The aim of this work is to detect and characterise out of plane wrinkles in complex glass-carbon structures using ultrasonic non-destructive tests. The parts to be inspected are made of epoxy resin reinforced with unidirectional carbon fibre and woven glass fibre. In order to determine the ideal probe frequency and bandwidth with which the inspections must be made, the resonant frequency of the plies was determined using a MATLAB-based analytical model. The parts were then inspected using full-matrix capture (FMC) and the total focusing method (TFM) as well as a commercial phased-array instrument providing data that can be inverted to give out-of-plane ply direction. Finally, the amplitude of the wrinkle was measured visually on the samples and compared with the ultrasonic measurements to determine the accuracy of the proposed method.

Original languageEnglish
Pages508-521
Number of pages14
Publication statusPublished - Dec 2016
Event55th Annual Conference of the British Institute of Non-Destructive Testing, NDT 2016 - Nottingham, United Kingdom
Duration: 12 Sep 201614 Sep 2016

Conference

Conference55th Annual Conference of the British Institute of Non-Destructive Testing, NDT 2016
CountryUnited Kingdom
CityNottingham
Period12/09/1614/09/16

Fingerprint

Dive into the research topics of 'Wrinkle characterisation from ultrasonic scans of composites'. Together they form a unique fingerprint.

Cite this