Andreas Boenisch, Kontroll Technik, Germany
Extended abstract of paper presented at BINDT Annual Conference 2002, Southport, 17 Sept. 2002
Eddy Current inspection of aircraft structures has been standard practice for some time. Thermographic inspection is used for particular applications. This paper describes development of the methods specifically for deployment by means of a mobile vehicle and scanning device. The application of the eddy current method for location of fasteners for subsequent ultrasonic inspection is also described.
The ROBAIR project was commenced in 2001 and is a European CRAFT project led by Sonatest. The other partners are: Horton Levi, Kontrolltechnik (KT), NDT Consultants, Zenon, Noemon, Technical University of Sofia, South Bank University and TWI.
The basis of the project was that the NDT methods would be deployed by a mobile vehicle and/or scanner as shown in Fig 1. The NDT methods need to be adapted to be deployed by the scanner and vehicle.
2. Eddy current requirements
2.1. Surface crack detection
Typically eddy current techniques for surface crack detection in aircraft are carried out with high frequency probes. Detection of cracks initiating from a fastener require a scan to be performed around the fastener. The fastener is most likely to be aluminium, but could be another material.
The difficulty of carrying out scans around fasteners is that the parent/fastener interface is in effect a large crack, and if the fastener is steel, then the magnetic properties are detected by the probe.
It is therefore essential that the probe must be moved directly around the centre of the rivet, so that the contribution of the eddy current signal from the fastener or its interface are kept constant.
2.2. Sub-surface crack detection and corrosion detection
Typical inspections for 2nd and 3rd layer cracks and corrosion are carried out with low frequency probes. These can be ring probes with a transparent centre to locate the fastener, or one with a solid ferrite core, which may be used to detect corrosion in areas away from fasteners. More rapid inspection of fastener lines can be carried out with a sliding probe.
In the case of the low frequency method, precise placement of the probe relative to the fastener is also needed, and the sliding probe needs to run directly on the centreline of the fasteners.
2.3. Modification to techniques for robotic inspection
The scanner on Robair is capable of X/Y/Z movement. Therefore the X/Y co-ordinates of the fasteners must be known so that the scanner can be correctly positioned for other movements. Experiments with a C-scan representation of the eddy current signal showed a convenient way of locating the fastener co-ordinates, and of making the inspections easier.
Fig 2 shows the initial scans carried out by Kontroll Technik (KT) to test this method. Fasteners, slots and corrosion defects can clearly be seen.
Fig 3 shows results from a thicker section sample. In this case the phase of the signal has been used to emphasise the slot signals. KT's analysis software also gives a set of co-ordinates of the cursor as shown. Modifications are in hand to output these co-ordinates to the scanner for probe positioning.
Thermal imaging cameras are frequently used for detection of ice or hot spots on aircraft, and attempts have been made to produce images using lock-in and pulsed thermography (for example) to detect delaminations in composites.
The initial proposal for Robair was to mount a small heat source on the robot vehicle, and to track this using a reflecting system to a high performance thermographic camera mounted on the ground. There was originally no possibility of mounting any system on the vehicle because of the weight. However, a lightweight camera was provided by Horton Levi and it was decided that this could also be mounted on the vehicle.
Two other problems remained, these were the nature of the heat source, and a suitable configuration. After some experimentation the arrangement shown in Figure 4 was arrived at. It was shown that the surface could effectively be viewed through a reflector. The camera could therefore be mounted on the vehicle directly. This arrangement also reduces the possibility of reflection from the aeroplane's surface of lights and heaters in the hanger.
The choice of heat source was more problematical, and various methods were used. Variables such as heating time and emissivity were also studied. A range of techniques to adapt to different situations was developed, and two examples of the results are given in Fig 5-6.