[2A1] Multi-sensor electromagnetic inspection feasibility for aerospace composite surface defects
E Mohseni¹, S Pierce¹, R Vithanage¹, G Dobie¹, C MacLeod¹, S McKnight¹, E Foster¹, C Loukas¹, M McInnes¹, H Gover¹, K Burnham², J O’Brien-O’Reilly³, S Paton³, G Munro³,T O’Hare⁴ and M Grosser⁵
¹University of Strathclyde, UK
²National Manufacturing Institute Scotland (NMIS), UK
³Spirit AeroSystems, UK
⁴Spirit AeroSystems Belfast, UK
⁵Spirit AeroSystems, USA
The UK’s presence at the forefront of composite manufacturing in Europe has never been more important given how vital these structures are for: (i) slowing climate change through the reduction of fuel consumption and carbon footprints in different industries; and (ii) the development of wind and tidal blades to generate cleaner energy to achieve the net-zero target by the middle of the century. Therefore, composite technology, carbon fibre-reinforced polymers (CFRP) in particular, has been dominating the aerospace, energy and defence industries and this trend is expected to grow in the years to come.
Non-destructive evaluation (NDE) is essential during manufacturing to identify any defects early in the process as, if defects remain undetected, they could have far-reaching implications for the cost of scrapped/repaired parts and the safety of final components and also at the later stages of manufacturing and post-manufacturing to ensure the quality, integrity and fitness-for-service of these safety-critical components. Although ultrasound testing (UT) has been predominantly used for the inspection of CFRPs owing to its excellent performance for bulk NDE inspections, the method is not sufficiently sensitive to all defect types occurring in such components. Ultrasonic waves transmitted using array probes on CFRP components mainly interact with defects that are extended perpendicularly to the direction of the wave propagation such as delamination. The technique does not offer sufficient sensitivity for the detection of shallow and narrow surface defects commonly created by matrix transversal cracking and barely visible impact damage mechanisms. The compound CFRP gives rise to the mixed electromagnetic properties where highly conductive carbon fibres are moulded in a dielectric resin matrix. This provides a unique opportunity to explore the potential of electromagnetic NDE sensing modalities such as eddy currents (EC) and electrical capacitance imaging (CI) for the inspection of surface defects.
Accordingly, this feasibility study was aimed at investigating the design, automated robotic delivery and performance assessment of different sensor technologies for the detection of surface defects through experiments. To this end, machined surface defects were fabricated in a CFRP sample. The automated robotic inspection was implemented for all UT, EC and CI sensors individually, where a novel sensor-enabled robotic system based on a real-time embedded controller was developed. The system components, consisting of a KUKA robotic arm, force/torque (F/T) sensor and NDE sensor and controller, were interfaced through a core program in LabVIEW, enabling: (a) real-time communication between different hardware; (b) data acquisition from all sensors; and (c) full control of the processes within the cell. Moreover, real-time robot motion corrections driven by the F/T sensor feedback were established to adjust the contact force and orientation of the sensors to the component surface during the scan. All sensors, including the UT roller probe, EC array and CI sensor boards, were robotically delivered on the designated surface notches with varying depths of 0.1, 0.2, 0.5 and 5 mm. The results of EC and CI testing showed enhanced detectability with a high signal-to-noise ratio (SNR) for the defects shallower than 0.2 mm when compared to the UT B-scan images.
Non-destructive evaluation (NDE) is essential during manufacturing to identify any defects early in the process as, if defects remain undetected, they could have far-reaching implications for the cost of scrapped/repaired parts and the safety of final components and also at the later stages of manufacturing and post-manufacturing to ensure the quality, integrity and fitness-for-service of these safety-critical components. Although ultrasound testing (UT) has been predominantly used for the inspection of CFRPs owing to its excellent performance for bulk NDE inspections, the method is not sufficiently sensitive to all defect types occurring in such components. Ultrasonic waves transmitted using array probes on CFRP components mainly interact with defects that are extended perpendicularly to the direction of the wave propagation such as delamination. The technique does not offer sufficient sensitivity for the detection of shallow and narrow surface defects commonly created by matrix transversal cracking and barely visible impact damage mechanisms. The compound CFRP gives rise to the mixed electromagnetic properties where highly conductive carbon fibres are moulded in a dielectric resin matrix. This provides a unique opportunity to explore the potential of electromagnetic NDE sensing modalities such as eddy currents (EC) and electrical capacitance imaging (CI) for the inspection of surface defects.
Accordingly, this feasibility study was aimed at investigating the design, automated robotic delivery and performance assessment of different sensor technologies for the detection of surface defects through experiments. To this end, machined surface defects were fabricated in a CFRP sample. The automated robotic inspection was implemented for all UT, EC and CI sensors individually, where a novel sensor-enabled robotic system based on a real-time embedded controller was developed. The system components, consisting of a KUKA robotic arm, force/torque (F/T) sensor and NDE sensor and controller, were interfaced through a core program in LabVIEW, enabling: (a) real-time communication between different hardware; (b) data acquisition from all sensors; and (c) full control of the processes within the cell. Moreover, real-time robot motion corrections driven by the F/T sensor feedback were established to adjust the contact force and orientation of the sensors to the component surface during the scan. All sensors, including the UT roller probe, EC array and CI sensor boards, were robotically delivered on the designated surface notches with varying depths of 0.1, 0.2, 0.5 and 5 mm. The results of EC and CI testing showed enhanced detectability with a high signal-to-noise ratio (SNR) for the defects shallower than 0.2 mm when compared to the UT B-scan images.
