Automating composites repair

Consistent with the drive towards greater automation in composites, GKN Aerospace is tackling one of the most labour intensive parts of the lifecycle of composites by enabling a mechanised route for the preparation of damaged parts to accept repair patches. Dr Neil Calder reports. Repair of structural defects from in service or manufacturing events has been the Achilles’ heel of composites applications right from the outset and this effort by GKN Aerospace and selective laser coating removal company SLCR reported at the Farnborough Airshow provides part of the solution to that perennial problem. The relationship between GKN and SLCR, introduced publically at the JEC show in March 2009, has been building for some time. SLCR is most noted for its work in surface treatment based on laser technology, in material removal, cleaning and preparation. Its applications include paint stripping and surface cleaning as well as artwork restoration. In composites, where a structural defect such as local delamination or inclusion exists some way down through the laminate, it can turn into quite a big repair, requiring grinding or abrading back layer by layer to produce a stepped scarf repair to allow the reinsertion of plies back in the right orientation to make that as good as it was before. The design of the repair method depends on how deep it is and where the best access is, so repairs tend to be on a case by case basis. Some repair can happen just as part of the manufacturing process, correcting inclusions, porosity or difficulty with bonding, whilst other instances are as a result of in service damage, which would be detected principally as local delamination through non-destructive testing. Laser lesions The laser repair preparation technique involves ablating the resin fraction of a composite part using a high energy laser beam and then mechanically removing the exposed fibres in a stepwise, incremental manner. In very much the same method as most composite parts are built up, the process effectively deconstructs a patch of the part layer-by-layer until the structural defect area has been exposed and then removed and the remaining material is prepared for a replacement scarf joint. The laser used in this application is a CO₂ TEA laser, producing a very short duration pulse of high energy in the far infrared region where typical resins are highly absorptive. This type of laser is typically used for surface engineering or etching where the workpiece material is removed or transformed across a small processing area rather than through an intensely focused spot such as in laser cutting. SLCR is already using similar systems for paint stripping and surface preparation of composites, although using lower power densities. The laser energy is transferred during the 100ns pulse into a surface layer of the resin about 10µm deep, requiring a typical 10 passes at around 50Hz to expose a single lamination of a UD layup. Because of the different ways in which the laser energy is absorbed by the resin and by the fibres in the workpiece, the resin can be completely removed whilst leaving the fibres exposed but otherwise undamaged. The varying process rate with absorptivity of the various layers of composites and incremental processing combine to provide a situation where the process is able to remove one layer of material without affecting others. There are many applications where high power lasers are now being used to remove or modify surface layers. It is in its selectivity that the laser process is useful. The laser has a long depth of field enabling processing through a 3cm deep working volume, potentially accessing genuine 3D areas within composite components for repair. This means that the laser beam delivery system for this process can use a galvanometer raster scanning head, optically somewhat similar to those employed by Virtek layup projection systems. Although the current machine uses geometrically simple circular material removal schemes, with this degree of control any shapes are potentially possible. The equipment at the GKN composites development centre has demonstrated that the process works, in that when you remove one layer you do not damage the next and when you layup a repair patch it creates a successful bond. The mechanical test results of lap sheer specimens for the laser prepared scarf joints are better than if you use a traditional hand grinding technique. High power microscopy shows that in the conventionally prepared repair abraded by hand, at the interface a lot of fibres are broken up and discontinuous. What this laser process does is ablate the resin, exposing the fibres which can then just be brushed away: the fibres that are left at the interface are very much intact and continuous. That makes it more difficult to initiate the bond failure, giving a consequent increase in bond strength to the repair patch. Reducing repair time GKN is aiming for a 50% reduction in the time to prepare a typical composite repair patch, although this is an estimate of a typical benefit which would be accessed as a consequence of manpower reduction on repetitive repair tasks. Whilst a skilled operator would take some tens of hours to effect a normal repair, the machine could take much of the routine process of sanding down through composite layers out of this equation and will produce a much more standardised result. GKN can only capitalise on this if one repair is very much like another though, which will bring with it an approach of standardisation in the same way that fasteners are configured. This technology potentially offers a set of ‘band aids’. The use of a laser to ablate material in the region of the damaged area is not necessarily energy efficient per se, but the approach is one where the overall effect is significantly more time efficient. To have a definite number for the laser processing rate is misleading as this needs to be compared with the entire repair process before a meaningful comparison can be made. This mechanisation of repair preparation would potentially enable integration with other aspects of the process, such as NDT and programming for repair. The vision of a totally mechanised structural component care process is made much more realisable now this part of the jigsaw has been demonstrated. Once the technique of repair preparation has been taken into the CNC domain, the next steps towards mechanisation will necessarily follow. The advantages of this laser composite repair process are that for certain types of repair these can be effected automatically bringing with it the characteristics of mechanised processes of consistency and reduced labour content. This sits within the whole composites automation spectrum and aims to reduce the lifecycle cost of highly engineered composites structures. www.gkn.com