Polymer matrix composite repair

Abstract

A procedure for repairing a polymer matrix composite component is provided. The procedure includes the steps of: providing a polymer matrix composite component having a site prepared for repair by removal of damaged or defective material; locating an uncured, polymer matrix composite repair patch at the site to re-build the component thereat; and curing the polymer matrix of the repair patch by heating the patch using eddy currents induced by one or more alternating current coils. The repair patch is without metallic additives, such that the repaired polymer matrix composite after the curing step is also without metallic additives in the vicinity of the repair patch.

Claims

1. A procedure for repairing a polymer matrix composite component, the procedure comprising: providing a polymer matrix composite component having a site prepared for repair by removal of damaged or defective material; locating an uncured, polymer matrix composite repair patch at the site to re-build the component thereat; and curing the polymer matrix of the repair patch by heating the patch using eddy currents induced by one or more alternating current coils, including a process of determining a desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field produced by the one or more alternating current coils, and using said desired maximum temperature and said desired penetration depth to determine corresponding values of amp-turns and frequency which are then applied to the, or each, coil in the curing process; wherein the repair patch is without metallic additives, such that the repaired polymer matrix composite after the curing step is also without metallic additives in the vicinity of the repair patch.

2. A procedure for repairing a polymer matrix composite component according to claim 1, wherein the composite of the repair patch has sufficient intrinsic electrical conductivity without such additives to allow all the eddy currents used to heat the patch to flow only through the composite.

3. A procedure for repairing a polymer matrix composite component according to claim 1, wherein pressure is applied to the repair patch during the curing.

4. A procedure for repairing a polymer matrix composite component according to claim 1, wherein the polymer matrix composite of the repair patch is the same material as the polymer matrix composite of the component.

5. A procedure for repairing a polymer matrix composite component according to claim 1, wherein the polymer matrix composite of the repair patch is a carbon fibre reinforced polymer matrix composite.

6. A procedure for repairing a polymer matrix composite component according to claim 1, further comprising monitoring the patch temperature during the induction heating and adjusting amp-turns applied to the, or each, coil to attain a desired maximum temperature of the patch.

7. A procedure for repairing a polymer matrix composite component according to claim 1, including a preliminary step of removing the damaged or defective material from the component to prepare the site.

8. A procedure for repairing a polymer matrix composite component according to claim 1, in which a look-up table is used to relate the desired maximum temperature and said desired penetration depth to corresponding values of amp-turns and frequency.

9. A procedure for repairing a polymer matrix composite component according to claim 8, in which the desired maximum temperature and said desired penetration depth are related to corresponding values of amp-turns and frequency by: (i) providing geometric properties of the patch, geometric properties of the coil, electrical properties of the patch, and thermal properties of the patch; (ii) determining a desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field of the alternating current coil; (iii) estimating values for the amp-turns and frequency of the alternating current to be applied to a coil: (iv) using the provided geometric properties of the patch and the coil, the provided electrical properties, and the estimated values for the amp-turns and frequency to predict an eddy current density distribution in the patch and a corresponding penetration depth of the electromagnetic field produced by the coil, (v) using the predicted eddy current density distribution, the provided geometric properties of the patch, and the provided thermal properties of the patch to predict the maximum temperature of the patch, and (vi) repeating steps (iii) to (v) for different estimated values for the amp-turns and frequency of the alternating current until the predicted penetration depth and the predicted maximum temperature converge on the desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field.

10. A method of determining values for amp-turns and frequency of an alternating current to be applied to an alternating current coil used for induction heating of a polymer matrix composite repair patch, the method comprising: (i) providing geometric properties of the patch, geometric properties of the coil, electrical properties of the patch, and thermal properties of the patch; (ii) determining a desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field of the alternating current coil; (iii) estimating values for the amp-turns and frequency of the alternating current to be applied to a coil: (iv) using the provided geometric properties of the patch and the coil, the provided electrical properties, and the estimated values for the amp-turns and frequency to predict an eddy current density distribution in the patch and a corresponding penetration depth of the electromagnetic field produced by the coil, (v) using the predicted eddy current density distribution, the provided geometric properties of the patch, and the provided thermal properties of the patch to predict the maximum temperature of the patch, and (vi) repeating steps (iii) to (v) for different estimated values for the amp-turns and frequency of the alternating current until the predicted penetration depth and the predicted maximum temperature converge on the desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field.

11. A method according to claim 10, in which the estimated values for the amp-turns and frequency of the alternating current whose predicted penetration depth and the predicted maximum temperature match the desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field are stored in a look-up table as a key-value pair.

12. A method according to claim 10, in which steps (i) to (vi) are repeated for one or more of: different geometric properties of the patch; different geometric properties of the coil; different electrical properties of the patch; different thermal properties of the patch.

13. A non-transitory computer-readable medium encoded with one or more look-up tables constructed by the method of claim 11.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Embodiments will now be described by way of example only with reference to the accompanying drawings, which are purely schematic and not to scale, and in which:

(2) FIG. 1 shows double-sided coils for heating from opposite sides of a composite repair patch;

(3) FIG. 2 shows a single-sided coil for heating from one side of a composite repair patch;

(4) FIG. 3 shows schematically electrical conductivities in different directions of a carbon fibre reinforced polymer matrix composite layer;

(5) FIG. 4 shows a process flow for a composite repair procedure;

(6) FIG. 5 shows a modelling approach for constructing a look-up table for determining values for amp-turns and frequency are applied to an AC coil in order to achieve a desired maximum temperature of a repair patch and a desired penetration depth of electromagnetic field;

(7) FIG. 6 shows surface contours of modelled current density distribution (amps per square millimetre) in a unidirectional Carbon Fibre reinforced polymer matrix Composite (CFC);

(8) FIG. 7 shows surface contours of modelled electric field strength (volts per metre) in the CFC of FIG. 6;

(9) FIG. 8 shows surface contours of modelled dissipated power (watts) in the CFC of FIG. 6;

(10) FIG. 9 shows surface contours of modelled temperature (degrees Celsius) in the CFC of FIG. 6; and

(11) FIG. 10 shows part of a look-up table correlating value of amp-turns to maximum temperature for a coil frequency of 336 kilohertz.

DETAILED DESCRIPTION

(12) The following detailed description concerns composite materials that can be cured by eddy current heating. In particular, it concerns any polymer matrix composite having an intrinsic electrical conductivity which is sufficient to allow eddy currents to flow through it and thereby heat and cure it. A particularly useful form of composite having such a property, however, is carbon fibre-reinforced polymer matrix composite (CFC). The following detailed description therefore refers mainly to CFCs, but in principle the present disclosure applies more widely to other suitable composites.

(13) A key to successful repair of CFCs is to understand their material properties and exploit them electromagnetically using specific current and frequency combinations.

(14) Conventional approaches to repair of CFCs discussed above typically use a patch repair technique in which some form of electrically conductive additive is introduced into the patch. In the present case, however, there is provided an electromagnetic repair approach which does not require the use of such additives. The approach determines a combination of amp-turns and frequency to be used for the, or each, induction coil based a desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field. This determination may be based on factors such as the dimensions of the CFC repair patch, the orientation of the CFC layers in the patch, the carbon fibre volume fraction, and the distance of the coil to the repair surface. This can eliminate a need for electrically conductive additives to be used in the repair patch. The repair patch can thus avoid affecting any lightning strike protection layout or electromagnetic shielding. It can also avoid undermining the mechanical properties of the parent structure due to overtemperatures and avoid use of non-standard materials with associated needs for additional testing and certification. Furthermore, the approach disclosed herein can reduce repair times and costs because it enables repairs to be carried out on-site and in situ, and with relatively simple equipment requirements.

(15) The electromagnetic heating system disclosed herein can have one or more induction coils to enable repairs having different areas, depths and shapes. For example, the system may have double-sided coils 101 as shown in FIG. 1 for heating from opposite sides of a patch 102, or may have a single-sided induction coil 201 as shown in FIG. 2 for heating from one side of the patch 102.

(16) The electrical conductivity of CFCs is anisotropic as illustrated in FIG. 3, which shows schematically a CFC layer 301 with the highest conductivity along the fibre direction x, and lower electrical conductivities in the transverse direction y and through-thickness direction z, respectively. For the purposes of illustration, the conductivity in the fibre direction may be typically 40000 siemens per metre, whilst in the transverse and through-thickness directions the conductivity may only be between 20 and 200 siemens per metre, and between 1 and 10 siemens per metre respectively.

(17) This anisotropic electrical conductivity means that the skin effect (i.e. the penetration depth of electromagnetic fields into the CFC) is determined based on an orthogonal relationship between the direction of the magnetic field and the direction of the induced current in the composite. Thus considering orthogonal axes x, y, z, and a CFC whose carbon fibres extend in the x direction (see FIG. 3), the penetration depth δ.sub.x measured along the fibre direction when the magnetic field is along the y (transverse) direction is given by:

(18) $\begin{array}{cc}{\delta }_x=\sqrt{\frac{2}{{\mathrm{\omega \mu }}_y{\sigma }_z}}& \left[\mathrm{Equation}1\right]\end{array}$
where ω is the angular frequency given by 2πf, f being the frequency of the alternating current applied to the coil, μ.sub.y is the magnetic permeability of the CFC in the y direction, and σ.sub.z is the electrical conductivity of the CFC in the z (through-thickness) direction.

(19) However, the magnetic permeability in a CFC is effectively constant in all directions and can also be treated as having the same magnetic permeability as air (μ.sub.0). Thus:

(20) $\begin{array}{cc}{\delta }_x=\sqrt{\frac{2}{{\mathrm{\omega \mu }}_0{\sigma }_z}}& \left[\mathrm{Equation}2\right]\end{array}$

(21) Similarly, the penetration depth δ.sub.z measured in the through-thickness direction z when the magnetic field is along the transverse y direction is given by:

(22) $\begin{array}{cc}{\delta }_z=\sqrt{\frac{2}{{\mathrm{\omega \mu }}_0{\sigma }_x}}& \left[\mathrm{Equation}3\right]\end{array}$
where σ.sub.x is the electrical conductivity of the CFC in the x direction.

(23) Unlike the magnetic permeability, the electrical conductivity and thus the penetration depth depends strongly of the orientation of the fibres. The skin effect can therefore be manipulated based on dimensions and the lay-up of the repair patch to ensure complete heating of the repair area. In particular, as the electrical conductivity of the carbon fibres, and the CFC more generally, cannot be altered if conductive additives are not incorporated in the repair patch, the parameter which effectively controls the skin effect is the supply frequency to the coil.

(24) As set out above, the repair procedure disclosed herein includes a process of determining the coil amp-turns NI and frequency f combination required to achieve the appropriate curing temperature of the CFC resin matrix through the thickness of the repair patch. The process flow of the procedure is shown schematically in FIG. 4.

(25) Firstly a damaged or defective region of a CFC component is obtained at step 401, and is prepared at step 402 by removing (e.g. by grinding out) the damaged or defective material, thereby forming a repair site. The geometry and material of a repair patch to re-build the component at the site can then be defined at step 403. In particular, the dimensions of the patch and its material properties, such as lay-up, resin, volume fraction of carbon fibre, can be defined. These material properties are preferably matched as closely as possible to those of the removed material. At step 404, the patch is applied at the repair site and a vacuum bag placed over the patch to apply pressure to the patch while it is cured.

(26) The dimensions of the patch also determine at step 405 to a significant extent the size (i.e. diameter) of the, or each, alternating current coil used for induction heating the patch.

(27) A desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field to achieve an appropriate cure of the patch are then determined, which in turn allows the values for amp-turns (and hence current I) and frequency f to be determined which are then applied to the coil(s).

(28) In the present embodiment, a pre-computed LUT appropriate for the dimensions of the patch and its material properties, and also appropriate for the size and stand-off of the coil(s), is used at step 406 to determine values for amp-turns (and hence current I) and frequency f which are then applied to the coil(s) based on the desired maximum temperature of the patch and the desired penetration depth of the electromagnetic field produced by the coil(s) in order to cure the resin matrix. The procedure to relate the desired maximum temperature of the patch and the desired penetration depth of the electromagnetic field to the values for amp-turns and frequency, which may be used to generate the LUT or may be performed on-line, will be described further with reference to FIG. 5.

(29) The values of current I and frequency f are set at step 407 on the power supply to the coil(s), and the temperature of the patch is monitored at step 408 in order to ensure that it undergoes a suitable cure cycle. If necessary the coil current I can be adjusted at step 409 to maintain the cycle. Generally, the coil frequency f is kept fixed, however, in order to maintain an unchanged penetration depth.

(30) The final result of the procedure obtained at step 410 is a repaired component in which the patch has been cured without use of any conductive additives.

(31) In the above procedure, a pre-computed LUT is preferably used to determine the values for amp-turns NI and frequency f which are applied to the coil. The LUT can be constructed using a validated 3D multi-physics model-based approach. The 3D modelling captures the electrical and thermal properties of each layer of the CFC, e.g. assuming homogenous properties per layer for all the layers of the composite structure to be repaired. This modelling approach is shown schematically in FIG. 5.

(32) More particularly, a first stage performed at step 501 is to define geometric properties of the patch and the, or each, coil. For the patch, these can include in particular its shape, dimensions, lay-up arrangement (i.e. position and number of layers and fibre direction), volume fraction of carbon fibres. For the coil these can include in particular its diameter, number of turns N, and stand-off from the patch. The geometric parameters of the patch also allow its electrical and thermal properties to be defined. Thus the electrical properties can include electrical conductivities in longitudinal and transverse directions of the fibres and through-thickness directions of the layers, and temperature coefficients of the electrical conductivities. The thermal properties can include thermal conductivities in longitudinal and transverse directions of the fibres and through-thickness directions of the layers, densities, heat capacities and heat transfer coefficients.

(33) Next, at step 502, estimated values of the frequency f.sub.0 and amp-turns NI.sub.0 needed to achieve a desired maximum temperature T.sub.target in the patch and penetration depth are selected. An electromagnetic (EM) analysis based on a finite element model (FEM) of the patch and coil(s) is then used to determine an eddy current density distribution induced by the application of f.sub.0 and NI.sub.0 to the coil(s). Associated with this distribution is a determination of the penetration depth of the EM field in the patch. As well as the values for f.sub.0 and NI.sub.0, the EM analysis requires the defined geometric properties of the patch and the coil, and the electrical properties of the patch. The analysis can have suitable boundary conditions in order to contain the FEM to a finite size.

(34) FIG. 6 shows predicted surface contours of eddy current density Jo induced by a circular coil from an example EM analysis. The current distribution is stretched along the x direction of the carbon fibres. This non-circular distribution is caused by the anisotropic electrical conductivity, which is highest in the carbon fibre direction. The low electrical resistance along the fibres means that the current flows closest to the applied magnetic field in the fibre direction. However, to complete a circuit, the current also has to cross high resistivity material in the though-thickness direction z. The current therefore spreads out to increase its cross-sectional area and thereby reduce the resistance, hence the overall oval shape of the eddy current distribution.

(35) Referring again to FIG. 5, an FEM-based thermal analysis is then performed at step 503 to predict the maximum temperature of the patch. The thermal analysis requires the defined geometric and thermal properties of the patch. Thermal boundary conditions can be set using the defined heat transfer coefficients. Conveniently, the eddy current density Jo can be converted into a dissipated power distribution P.sub.eddy for inputting into the thermal analysis.

(36) More particularly, FIG. 7 shows the predicted surface contours of electric field strength E.sub.0 for the EM analysis of FIG. 6, the electric field distribution being related directly to the eddy current density by the expression:
J=σE  [Equation 4]

(37) The dissipated power distribution P.sub.eddy from Joule heating is determined by the expression:

(38) $\begin{array}{cc}{P}_{\mathrm{eddy}}=\int E.Math.J\mathrm{dv}=\int \frac{J^2}{\sigma }\mathrm{dv}& \left[\mathrm{Equation}5\right]\end{array}$

(39) FIG. 8 shows the predicted surface contours of dissipated power distribution P.sub.eddy for the EM analysis of FIG. 6.

(40) FIG. 9 then shows the predicted surface contours of temperature from the thermal analysis using the dissipated power distribution P.sub.eddy of FIG. 8. The temperature distribution follows, as expected, a pattern to P.sub.eddy of FIG. 8. This illustrates the impact of the anisotropic thermal properties of the CFC. In particular, due to the large anisotropic changes in electrical and thermal conductivity, heat is mostly generated where the current density is lowest (FIG. 6) because of the low electrical conductivity to current flow orthogonal to the fibre direction.

(41) Returning to FIG. 5, the modelling can include an automated return loop comprising steps 504 and 505 to update the values of f.sub.0 and NI.sub.0 if the maximum temperature T.sub.0 predicted in the patch does not match the desired maximum temperature T.sub.target.

(42) By repeating the modelling of FIG. 5 for different pairs of values for f.sub.0 and NI.sub.0, an LUT can be constructed from the pairs of values for f.sub.0 and NI.sub.0 and the predicted penetration depth and predicted T.sub.0 to which each pair is related. FIG. 10 shows an example of part of such an LUT, and correlates value of NI.sub.0 to T.sub.0 for an f.sub.0 of 336 kilohertz.

(43) When using the LUT in the repair procedure of FIG. 4, one option is to convert the table into the form of an empirical equation expressing amp-turns and frequency in terms of penetration depth and maximum temperature. Another option, however, is simply to interpolate amp-turns and frequency values as needed directly from the data of the table.

(44) Overall, the LUT takes complex model parameters and simplifies them into a form where basic model inputs can be converted into the required current and frequency outputs to achieve a particular cure temperature for a given set of geometric properties of the patch and coil. Typically a given repair shop will repair composites over a restricted range of geometries, and thus will require access to only a limited number of LUTs to cover their range of activities.

(45) It will be appreciated, however, by those skilled in the art that as an alternative to producing an LUT, the determination of the values of amp-turns and frequency on the basis of determining a desired maximum temperature of the patch and a desired penetration depth of the electromagnetic field may be performed at the stage of curing the patch.

(46) As well as advantages already mentioned above, the repair procedure:

(47) is efficient compared to conventional repair techniques such as thermal blanket and autoclave approaches;

(48) improves heat distribution in through-thickness repairs;

(49) provides a temperature feedback loop for control;

(50) is applicable to existing, certified CFC materials; and

(51) focuses heat on a local area, eliminating or reducing damage to parent structures.

(52) In aerospace applications, the procedure may be used to repair e.g. composite integrated accessory rafts, fan cases, fan blades and airframe components. However, the procedure may use in other sectors such as marine (e.g. propeller blades, deck machinery, azimuthing thrusters, etc.) and automotive (e.g. motorsport, lightweight roadgoing vehicles, etc.).

(53) Embodiments may be described as a process which is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be re-arranged. A process is terminated when its operations are completed, but could have additional steps not included in the figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination corresponds to a return of the function to the calling function or the main function.

(54) While the disclosure has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the disclosure set forth above are considered to be illustrative and not limiting. Moreover, in determining extent of protection, due account shall be taken of any element which is equivalent to an element specified in the claims. Various changes to the described embodiments may be made without departing from the spirit and scope of the disclosure.

(55) All references mentioned herein are hereby incorporated by reference.