WO2011131995A1

WO2011131995A1 - Testing joints between composite and metal parts

Info

Publication number: WO2011131995A1
Application number: PCT/GB2011/050803
Authority: WO
Inventors: Jonathan Meyer
Original assignee: Eads Uk Limited
Priority date: 2010-04-22
Filing date: 2011-04-21
Publication date: 2011-10-27
Also published as: GB2479776B; GB2479776A; GB201006731D0

Abstract

It is difficult to make a reliable connection between metal parts and fibre-reinforced plastics parts, as is needed for instance in the aerospace industry. The Hypin system, in which metal protrusions (5) extend from the metal part 1 into the plastics part (18), is promising, but it is difficult to test non-destructively for an adequate joint, i.e. with the pins (5) surrounded by fibres, with no voids or resin-rich areas. To solve this problem,a set of metal terminals (42) is inserted or embedded in the component (18) around the joint, so as to pass into the fibre reinforcement. Conductance measurements made between the metal part (1) and the terminals42 indicates how well the metal protrusions extend into the CFRP sheet, contacting the fibre layers (15).

Description

TESTING JOINTS BETWEEN COMPOSITE AND METAL PARTS

FIELD OF THE INVENTION

The present invention relates to a method of inspecting a joint between two components; it has particular application to a hybrid penetrative joint between a metal component and a fibre-composite component.

BACKGROUND OF THE INVENTION

Joining between composite and metallic or thermoplastic components is currently approached in a number of ways, each with its own limitations. The use of fasteners is commonplace but tends to result in de-lamination around fastener holes, as well as the

associated difficulties of drilling holes in composites containing reinforcement such as Carbon Fibre and Aramids such as Kevlar. The bearing strength of laminated composites tends to be low, as does the inter-laminar shear strength. This results in a requirement for significant reinforcement around fastener holes, leading to a large weight increase, which is particularly undesirable in aerospace

applications.

Adhesive bonds are an increasingly common means of joining metallic components to composite laminates, however these perform poorly in peel, tension and cleavage, and tend to fail with little or no warning. Their weakness in peel and in tension makes bonded joints similarly limited in their application within conventional aerospace structures. Any mitigation for the poor performance in peel or tension tends to result in large bond surface areas, with the associated weight penalties that go with this. Existing research into the use of surface features to improve the strength of metallic/composite joints is limited.

WO 2008/110835 (Airbus) provides a method of joining a pair of components, in which an array of projections is grown on a bond region of a first (metal) component by stacking a series of layers, to form a kind of spiky mat on the first component. The latter is then urged into a second (CFRP) component so that the projections are embedded in the matrix of the second component. An intimate bond is thus formed. This is known internally as a Hybrid Penetrative Reinforcement (Hyper) Joint, and the projections are known as Hypins . WO 2008/110835 discusses particular methods of making such pins, but the present invention is not limited to these: another possibility is shown in WO 2004/08731 (TWI), for instance.

The Hyper Joint provides excellent mechanical properties, but, because of its complex geometry and large differential material densities in the interface region, it can be difficult to examine the joint for defects in a cost- effective manner. The most critical defects are:

1) Lack of fusion between the Hypins and the metallic part, and

2) Lack of fibre consolidation around the Hypins

(resin-rich pockets) SUMMARY OF THE INVENTION

According to an aspect of the invention there is provided a method of testing the integrity of a joint of the Hypin type between a metal or other conductive part and a fibre- composite part, as defined in claim 1. The invention also relates to assemblies incorporating such a joint, and to suitable test rigs.

The fibre-composite part will also be adequately

electrically conductive, in most cases by virtue of the carbon fibres used as reinforcement. The resistance between the metal part and a point, preferably several points, in the fibre-composite part is measured, and this measurement is compared with a predetermined standard. The measurement points are formed by conductive pins extending into the thickness of the composite part so as to make contact with a number of layers of fibre reinforcement, and this gives an indication of how well the projections

("Hypins") contact the conductive fibre layers.

Electrical methods for measuring joint integrity are of course known, as for instance from WO 2004/079353 (Pass Tecnologies) , which uses conductive patterns to check a seal, or US 3916304 (University of Akron/ Roemer et al . ) where welds are tested by current probes. However, such methods are not obviously applicable to hybrid joints. Embodiments of the invention use a sensitive ohm-meter or similar device to measure the resistance between a known location in the laminate, such as a pin-hole drilled to expose fibres, or a measurement terminal embedded in the composite, and a location on the metallic part. The resistance will increase if there are significant resin- rich areas between the Hypins and the conductive fibres in the composite, or a significant lack of fusion between the pins and the base part. The variation can be expected to be in the region of mQ or even μΩ. A method of this type clearly only works for composites utilising electrically conductive fibres (e.g. carbon) and joints utilising conductive materials (e.g. steel, titanium), such materials being referred to for convenience as metals.

Without this method of non-destructive examination (NDE) , the alternatives would be limited to X-ray methods such as CT, which are extremely expensive and difficult to

implement for large hybrid structures. The invention allows an initial identification of suspect joints, that can then be passed on for examination by such more

expensive methods. Preferably there are several measurement terminals or points, distributed around or along the joint. In

conjunction with computer modelling, these multiple

reference locations can be used to build a ^xpicture' of where the defects in the joint are located. Such a method takes advantage that of the fact that the fibre- reinforcement layers are generally directional and laid down in a criss-cross or alternating way. An unacceptably low conductance measurement at a pair of measurement pins can thus be used to indicate inadequate bonding of a Hypin located at the intersection of two fibre directions from the pair of measurement pins.

This method is very low in cost, and could be applied in an in-service as well as a production environment. For example, an assembly could have permanently embedded sensors associated with a joint to perform a structural- health-monitoring role, detecting damage propagation due to fatigue or in-service incidents. It is also very portable, and can be used on parts of very large scale where other NDT/NDE methods would be difficult to implement.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the invention, embodiments will now be described with reference to the accompanying drawings, in which:

Figure 1 is a cross-sectional view of a Hyper Joint, with measuring apparatus indicated schematically;

Figure 2 is a simplified view in plan of a joint,

showing an array of measuring pins; and

Figure 3 is a graph of a measurement set including

unacceptable values. DETAILED DESCRIPTION

Figure 1 shows a joint between a metallic part, for

instance a floating rib foot 1, and a CFRP part 18 in the form of a sheet, for instance a wing cover. In such a sheet there will usually be several carbon-fibre layers or mats 15, each containing fibres extending in substantially the same direction, the layers alternating at angles such as 0°, 90°, 45° and 135°, as indicated schematically by dashed lines of differing length. The carbon fibres are embedded in epoxy resin.

The metallic part comprises a web portion 3 forming an upstanding bracket part and a pair of flanges 2 forming a base. An array of projections or joint pins ("Hypins") 5 extends from the underside of the flanges 2. As can be seen in Figure 1, the projections 5 are distributed more or less evenly, in a regular array, over a bond region which extends over most of the area of the flanges 2.

The base 2 lies flush with a CFRP sheet 18, with the projections 5 penetrating the sheet to approximately half to two-thirds its depth, and passing through at least two carbon-fibre mats 15. The projections should be long enough to penetrate several of the carbon-fibre mats.

Typically the projections would be 3-4 mm long and the CFRP sheet would be perhaps 5-6 mm thick, though if it is thicker the projections need be no longer. The projections preferably have arrow-heads as shown, and ideally these physically engage behind the carbon fibres.

In fact, a sheet of glass ply 19 is inserted between the base 2 and the CFRP part 18 and is also penetrated by the projections; this will be described later. To assemble the parts, the projections are pressed into the prepreg, or the layup, so that they are embedded in the laminate,

penetrating the carbon-fibre layers, and the assembly is then cured. The result is usually a good join. However, it cannot be inspected by eye, and there could be voids around the pins, or resin regions inadequately penetrated by carbon fibres. Such points will give an inadequate electrical contact between the projection and the carbon fibres.

To test the joint, therefore, the following procedure is adopted .

1) A component part is provided, consisting of a

metallic part or bracket 1 attached to a Carbon-Fibre panel by means of a Hyper joint (as described in

WO 2008/110835) . A valuable innovation is the addition of a single ply of glass-fibre material 19 adjacent to the interface with the bracket to act as an electrical insulator.

2) Either:

a single hole or a plurality of small holes 40 is drilled in the laminate at a known distance (s) from the metallic part interface, or

a single pin 42 or a plurality of small conductive, preferably metallic, pins is embedded in the laminate with their surfaces co-planar with or protruding above the composite surface. In principle, both these measures could be adopted. The pins or holes could typically be 3-5 mm long and 1 mm in diameter.

3) A sensitive device 100 for electrical resistance or conductive measurement is connected at one end 102 to a known location on the metallic part, and at the other end 104 to either a pin inserted into the drilled hole or a terminal pin that is made to contact an embedded pin at the laminate surface.

4) The measured resistance between the bracket 1 and the pin or pins is a function of the distance from the interface and the connectivity between the Hypins and the carbon fibres in the laminate. The insulating ply 19 eliminates conduction between the metallic part face 2 and the laminate 18, and so the only conductive path is through the pins.

5) By calibration through test articles with seeded

defects and/or simulation using computational

numerical methods it is possible to identify the presence and location of a defect by measuring the resistance between multiple locations and comparing the benchmark values.

The pins 42 (or the holes 40) penetrate the insulating panel 18 to about the same depth as the projections 5, preferably slightly more, so that there is a conducting path from the pins 42 along the conducting fibres 15 of each layer to the projections 5, provided that the

projections contact the fibres at that level and are not insulated by a void or by inadequate fibre coverage.

In general, the principle is that the plurality of current paths represented by the Hypins behave as parallel

circuits, and so the anticipated total resistance, or rather conductance, measurement would be proportional to the sum of the individual conductance at each pin, as follows :

1/Rtotal = (1/Rl) + (1/R₂) + (1 s) +-... (1/Rn)

This formula is the basis for identifying the effect of a defect, albeit with the complication that there will be some variability due to the distance between the pins and the measuring points. However, it is this variability in distance that makes it possible to pinpoint the location of a defect if multiple measurements are taken between

different locations. The test setup comprises the resistance measurement device, a plurality of pins embedded at known locations around the conductive part, and a testing rig with contacts designed to locate on the embedded pins and the conductive part in known locations. Theoretically a smaller number of

contacts could be used, designed to be movable between the embedded pins on the CFRP part. A small microprocessor is used to analyse the incoming currents and calculate whether a defect is present and whether it is in a critical region (typically the edges of the joint are more critical as they carry more load) . This could then simply give a green or red light as an indication to the user as to whether further investigation is required.

The way defects can be located can be understood by

reference to Figure 2. A bond region between the flange 2 of the metallic part and the fibre-reinforced plastics is shown as a square, but this is purely for simplicity, and any shape is conceivable. A plurality of conductive pins 42 is embedded in the plastics part so as to surround the bond area, or at least the accessible part of it, and can be contacted for making measurements of conductivity between them and the metal part. The pins are laid out in an extension of the same grid that the Hypins 5 are arrayed in, because this allows a simple method of locating

defective contacts. However, this is not inevitable, though it is advantageous if the spacings are comparable.

In this example, two of the measurement pins 42 exhibit unacceptably high resistance readings. The measurements are shown schematically in Figure 3, discussed below.

Notional lines drawn from these in the known direction of the conductive fibres intersect at three points on the grid of the Hypins, shown encircled. The "acceptable" lines are shown dashed, the "unacceptable", or at least "suspect", chain-dotted. Remedial action can then be taken at these pins, or the process inspected to see if there is a

systematic reason for the failures.

Figure 3 shows a typical set of measurements at each of the pins. "Acceptable" measurements lie in an envelope E, which while not continuous has a generally continuous range of values. The two unacceptable readings B lie outside the envelope. These values and ranges would generally be determined empirically by examination of joints known to be acceptable and joints having specific defects; the

acceptable conductivity measurements will depend on such factors as the fibre density, the size of the pins, the layout, the shape of the joint and so on.

In the foregoing, reference is made to a CFRP part, but the invention is applicable to any fibre-reinforced part, provided the fibres have adequate conductivity. Also, the protrusions can be made by any suitable method, not merely that described in the earlier Airbus application. The invention is also applicable to a joint of the kind shown in Figure 10 of WO 2008/110835, in which a metal "mat" of protruding pins is sandwiched between two CFRP layers.

Here there would be measurement pins embedded in each of the CFRP layers. The grid arrangement is shown as square, but other possibilities can be contemplated.

Claims

1. A method of testing a joint between a conductive part (1) and a conductive-fibre-reinforced plastics

component (18), the conductive part (1) having a number of protrusions (5) extending into and retained in the plastics component (18),

in which electrical contact to the plastics component is made via a metal terminal (42) inserted or embedded in the component so as to make contact with the fibre reinforcement, and the electrical resistance or

conductance between the conductive part and the plastics component is compared with a predetermined value.

2. A method according to claim 1, in which contact is made to the fibre-reinforced component by drilling a hole (40) into it and inserting a conductive pin (42) as the metal terminal .

3. A method according to claim 1, in which the component

already contains an embedded metal terminal, to which contact is made.

4. A method according to any preceding claim, in which there are two or more pins embedded in the plastics component, and two or more comparisons are made.

5. A method according to claim 4, in which the metal

terminals are provided so as at least partly to surround the joint.

6. A method according to claim 4 or 5, in which the location of any defect is found using the two or more

measurements .

7. A method according to any preceding claim, in which the conductive-fibre-reinforced plastics component (18) contains two or more layers of fibres, the fibres in each layer extending in substantially the same direction and the fibre directions in adjacent layers differing by at least 45 ° .

8. A method according to claim 7, in which the metal

terminals and the protrusions penetrate at least half the fibre layers.

9. A method according to any preceding claim, in which there are two fibre-reinforced plastics components, with the conductive part being planar and sandwiched between the plastics components.

10. An assembly comprising a conductive part (1) fixed to a conductive-fibre-reinforced plastics component (18) by a number of protrusions (5) extending from the

conductive part (1) into the plastics component (18) so as to penetrate the fibres, the assembly further

including a set of metal terminals (42) inserted or embedded in the component (18) so as to pass into the fibre reinforcement.

11. An assembly according to claim 10, in which the

conductive-fibre-reinforced plastics component (18) contains two or more layers of fibres, the fibres in each layer extending in substantially the same direction and the fibre directions in adjacent layers differing by at least 45 ° .

12. An assembly according to claim 10 or 11, in which a layer of glass-fibre material (19) is provided between the conductive part and the plastics component.

13. A test rig for testing a joint between a conductive

part (1) and a fibre-reinforced plastics part (18), including an apparatus (100) for measuring electrical resistance, connected between the metallic part and the plastics part, the connection to the latter being made by way of a pin (42) embedded in the plastics part.

14. A test rig according to claim 9 for testing parts in which several such pins (42) are embedded, the rig having a corresponding set of terminals adapted to connect to the pins, and a processor for evaluating the results of the resistance measurements.

1 5. A test rig according to claim 14, in which the processor is adapted to locate a faulty point in the joint by comparing the locations of pins giving low readings with the known orientation of the conductive fibres in the plastics part.