JP2020139747A - Probe and detection method of fiber orientation defect using the same - Google Patents

Abstract

To provide a probe capable of detecting orientation deviation with high accuracy of about 1° for an orientation deviation angle of ± 3° or more, and a swell angle of fiber swelling of ± 2° or more in conductive fibers in the cured layer constituting the conductive fiber reinforced laminate.SOLUTION: A uniform excitation differential probe P is composed of a uniform excitation coil 10 formed in a tubular shape by spirally winding a lead wire, and detection sensors 11a to 11n arranged along one surface of the uniform excitation coil 10 and in a row in a direction orthogonal to the extending direction of the lead wire on the one surface, in which two or more, and any two of which are used for detecting the difference in induced electromotive force caused by the eddy current U formed in a laminated body 1 by the excitation of the uniform exciting coil 10.SELECTED DRAWING: Figure 3

Description

According to the present invention, a macroscopic misalignment of a fiber-reinforced laminated body using conductive reinforcing fibers such as carbon fiber and metal fiber as a reinforcing material (angle in the orientation direction of the reinforcing fibers of each cured layer constituting the laminated body). The present invention relates to a probe for detecting misalignment (deviation) and local misalignment (fiber waviness of the cured layer) with high accuracy, and a method for detecting misalignment using the probe.

Currently, carbon fiber reinforced plastics (hereinafter referred to as CFRP) is one of the widely used fiber reinforced laminates. CFRP is used in various products such as golf club shafts, tennis racket frames, fishing rods, and laptop housings due to its characteristics of light weight, high strength, and high rigidity. Recently, it has begun to be used in aircraft fuselage and wings, propellers for wind power generators, racing car bodies, and bicycle frames.

The prepreg used in producing CFRP is formed into a sheet by aligning carbon fibers, which are fibrous reinforcing materials, in one direction or knitting them as a woven fabric, and thermosetting by mixing them with a curing agent and other additives. It is uniformly impregnated with a sex resin (for example, an epoxy resin) or a thermoplastic resin. The uncured one is called "A stage", the semi-cured one is called "B stage", and the cured one is called "C stage".
The CFRP laminate is obtained by laminating a plurality of "B stage" prepregs and heating and pressurizing them to cure them. The orientation direction of the carbon fibers and the number of laminated carbon fibers are appropriately selected according to the required mechanical properties of the members used. Each layer of the cured prepreg constituting the CFRP laminate is referred to as a cured layer.

In such a laminate, if the fiber directions of each cured layer are correctly arranged, the mechanical properties as designed are exhibited, but the planned orientation direction of the carbon fibers of each cured layer at the time of design (this is the reference orientation). If the orientation angle of the carbon fibers in any of the cured layers of the actual CFRP laminate deviates from the reference orientation direction with respect to the direction and the symbol L), the entire CFRP laminate will have the mechanical properties as designed. I can't hope. The angular deviation of the fiber orientation of the cured layer is one of the fiber orientation defects and is called macroscopic misalignment.

The laminated prepreg is produced by impregnating carbon fibers aligned in one direction or carbon fibers woven as a woven fabric as described above with resin, but the arrangement of carbon fibers is coarse and dense, and the resin impregnation In the process, resin flow may occur, causing partial fiber turbulence (fiber swell).
Further, when the CFRP laminate that has been accurately laminated is cured, it is pressurized and heated. At this time, depending on the shape of the CFRP laminate, pressure is partially applied, which is combined with the melting of the resin in the portion. The carbon fibers in that part may be extruded to cause partial fiber waviness. This is also one of the fiber alignment defects and is called local misalignment.

If the carbon fibers of the laminate do not have the above macroscopic misalignment or local misalignment, the cured CFRP laminate exhibits the required mechanical properties, but in the presence of these, the mechanical properties are significantly reduced. According to past data, for example, if there is a macroscopic misalignment of 3 ° with respect to the alignment direction of the carbon fibers, the rigidity is said to decrease by about 10%, and the rigidity decreases significantly as the orientation angle increases. On the other hand, when the deviation of the local misalignment becomes about 2 °, the intensity decreases by about 40%. In other words, its mechanical strength drops to 60%.

Therefore, in order to ensure the mechanical properties of the CFRP laminate, an attempt was made to detect carbon fiber misalignment by an eddy current test, and a circular absolute value probe or a circular TR probe in which a circular exciting coil and a detection coil were separated. Was used. As a result, it was possible to obtain an image (contour diagram) of a striped pattern or a lattice pattern having a width of several mm along the orientation direction of the carbon fibers.
The circular absolute value type probe succeeded in detecting the fiber orientation of the two cured layers laminated on the upper and lower sides by the image of the coarse lattice pattern, and the circular TR type probe succeeded in detecting the fiber waviness by the coarse image.

However, these probes were originally devised for metal materials, and since magnetic field lines are emitted radially evenly in all directions around the exciting coil, the orientation direction of the laminated carbon fibers and the fibers Although swell could be detected as a tendency, no further detection was possible.

On the other hand, if X-rays of Patent Document 1 are used, fiber orientation and fiber waviness can be accurately measured.

Japanese Patent No. 6321878

G. Mook, et al., Composites Science and Technology, 61 (2001) 865-873 H, Heuer, et al., Composites Part B, 77 (2015) 494-501 C. Schmidt, et al., Composites: Part B, 56 (2014) 109-116

However, the X-ray transmission test method requires that the object be installed between the X-ray generator and the film, and there are various restrictions such as the shape, size, location, and consideration of the health of the worker. There was, and it could not be used easily.

Currently, CFRP laminates are used as large-scale structures in various applications as described above, and reliability in terms of mechanical properties is required. When the deviation of the orientation angle of the carbon fibers and the swell angle in the fiber swell become large as described above, the mechanical properties are suddenly impaired. As a desired value, in the case of macroscopic misalignment, the orientation deviation angle is required to be ± 3 ° or less. However, it is impossible to inspect with such high accuracy with a conventional probe, and at present, there is no inspection method that can guarantee the mechanical properties of the CFRP laminate.

The present invention has been made in view of the above-mentioned problems of the prior art, and the alignment angle of the carbon fibers in the cured layer, the deviation of the orientation angle of ± 3 ° or more in the fiber swell, and the fiber with the swell angle of ± 2 ° or more The subject is to provide a probe capable of detecting waviness with an accuracy of about 1 ° and a method for detecting the fiber orientation defect with high accuracy.

The invention according to claim 1 (uniform excitation differential probe P; FIGS. 3 to 6)
A uniform exciting coil 10 in which a conducting wire is spirally wound to form a tubular shape,
Any two or more of the uniform exciting coils 10 arranged along one surface and arranged in a row in a direction orthogonal to the extending direction of the conducting wire on the one surface. The two are characterized by being composed of detection sensors 11a to 11n used for detecting the difference in the induced electromotive force due to the excitation of the uniformly exciting coil 10.

The invention according to claim 2 (FIGS. 3 (c) (d), 4 (c) (d), 5 (c) to (f), 6 (a) (b)) is claimed 1. In the uniform excitation differential probe P described in
The detection sensors 11a to 11n are arranged outside or inside the uniform exciting coil 10.

The invention according to claim 3 (FIG. 4) is the uniform excitation differential probe P according to claim 1 or 2.
The winding start portion 10a and the winding end portion 10b of the lead wire of the uniform exciting coil 10 are wound more densely than the intermediate portion 10c thereof.

The invention according to claim 4 (FIGS. 5 (c) and 5 (d)) is the uniform excitation differential probe P according to claim 1 or 2.
The uniform exciting coil 10 is composed of a winding start portion 10a and a winding end portion 10b in the same winding direction without an intermediate portion 10c, and two or more detection sensors 11a to 11n are installed in the intermediate portion 10c. It is characterized by that.

The invention according to claim 5 (FIGS. 5 (e) and 5 (f)) is the uniform excitation differential probe P according to claim 1 or 2.
The uniform exciting coil 10 is composed of a winding start portion 10a and a winding end portion 10b in the same winding direction without an intermediate portion 10c, and coincides with the winding start portion 10a and the winding end portion 10b, and is inside or outside the winding start portion 10a. The detection sensors 11a to 11n are respectively installed.

6. The uniform excitation differential probe P according to any one of claims 1 to 5.
The exciting surface 10r of the uniform exciting coil 10 arranged so as to face the inspection surface 1a of the inspection object is formed to match the shape of the inspection surface 1a of the inspection object.

A seventh aspect of the present invention is the first method for detecting the alignment deviation angle Δθ (macroscopic misalignment) of the reinforcing fiber 2 of the present invention (FIGS. 1, 9, and 10).
The fiber direction of any one of the cured layers 4a of the fiber-reinforced laminated body 1 in which the plurality of cured layers 4a to 4n are laminated is set as the overall reference direction L of the laminated body 1.
The fiber direction determined in advance with respect to the reference direction L is defined as the inspection orientation direction Lb to Ln of the remaining cured layers 4b to 4n.
The line passing through the center of the selected pair of detection sensors 11a and 11b of the uniformly excited differential probe P according to any one of claims 1 to 5 is defined as the center line CL.
The center line CL is installed on the inspection surface 1a side of the laminated body 1 in accordance with the inspection orientation direction Li of the target cured layer 4i, and the difference is the angle between the reference direction L and the center line CL. While maintaining the moving angle Ψi, the probe P is moved while supplying a high frequency current to the exciting coil 10.
The induced electromotive force generated in the pair of detection sensors 11a and 11b by the eddy current U generated by the reinforcing fiber 2 of the target cured layer 4i is converted into an electric signal, respectively.
The difference between the two converted electric signals (analog or digital signal) is calculated, and the orientation deviation angle Δθ of the reinforcing fiber 2 of the hardened layer 4i, which is the target with respect to the inspection orientation direction Li, is calculated based on the difference value. Calculate and
This is a method for detecting a fiber orientation defect of a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the orientation deviation angle Δθ exceeds ± 3 °.

Claim 8 is a second method for detecting the alignment deviation angle Δθ (macroscopic misalignment) of the reinforcing fiber 2 of the present invention (FIGS. 1, 9, and 10).
One reference direction L is set for the fiber-reinforced laminated body 1 in which a plurality of cured layers 4a to 4n are laminated.
The fiber directions of each of the cured layers 4a to 4n, which are predetermined with respect to the reference direction L, are set as the inspection orientation directions La to Ln.
Orientation direction for inspection of the cured layer 4i targeting the center line CL passing through the center of the selected pair of detection sensors 11a and 11b of the uniformly excited differential probe P according to any one of claims 1 to 7. A high-frequency current is applied to the exciting coil 10 while being installed on the inspection surface 1a side of the laminated body 1 in accordance with Li and maintaining a differential angle Ψi which is an angle between the reference direction L and the center line CL. The probe P is moved while supplying
The induced electromotive force generated in the pair of detection sensors 11a and 11b by the eddy current U generated by the reinforcing fiber 2 of the target cured layer 4i is converted into an electric signal, respectively.
The difference between the two converted electric signals (analog or digital signal) is calculated, and the orientation deviation angle Δθ of the reinforcing fiber 2 of the hardened layer 4i, which is the target with respect to the inspection orientation direction Li, is calculated based on the difference value. Calculate and
This is a method for detecting a fiber orientation defect of a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the orientation deviation angle Δθ exceeds ± 3 °.

A ninth aspect of the present invention is the first method for detecting the swell angle φ (local misalignment) of the fiber swell of the reinforcing fiber 2 of the present invention (FIGS. 1, 11, and 12).
The fiber direction of any one of the cured layers 4a of the fiber-reinforced laminated body 1 in which the plurality of cured layers 4a to 4n are laminated is set as the overall reference direction L of the laminated body 1.
The fiber directions of the remaining cured layers 4b to 4n, which are predetermined with respect to the reference direction L, are set as the inspection orientation directions Lb to Ln.
Orientation direction for inspection of the cured layer 4i targeting the center line CL passing through the center of the selected pair of detection sensors 11a and 11b of the uniformly excited differential probe P according to any one of claims 1 to 7. A high frequency is applied to the exciting coil 10 while being installed on the inspection surface 1a side of the laminated body 1 so as to be orthogonal to Li and maintaining a differential angle Ψi which is an angle between the reference direction L and the center line CL. The probe P is moved while supplying an electric current,
The induced electromotive force generated in the pair of detection sensors 11a and 11b by the eddy current U generated by the reinforcing fiber 2 of the target cured layer 4i is converted into an electric signal, respectively.
The difference between the two converted electric signals (analog or digital signal) is calculated, and the swell angle φ of the fiber swell 8 portion of the reinforcing fiber 2 of the target cured layer 4i is calculated based on the difference value. ,
This is a method for detecting a fiber orientation defect of a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the waviness angle φ exceeds ± 2 °.

A tenth aspect of the present invention is a second method for detecting the swell angle φ (local misalignment) of the fiber swell 8 of the reinforcing fiber 2 of the present invention (FIGS. 1, 11, and 12).
One reference direction L is set for the fiber-reinforced laminated body 1 in which a plurality of cured layers 4a to 4n are laminated.
The fiber directions of each of the cured layers 4a to 4n, which are predetermined with respect to the reference direction L, are set as the inspection orientation directions La to Ln.
Orientation direction for inspection of the cured layer 4i targeting the center line CL passing through the center of the selected pair of detection sensors 11a and 11b of the uniformly excited differential probe P according to any one of claims 1 to 7. A high frequency is applied to the exciting coil 10 while being installed on the inspection surface 1a side of the laminated body 1 so as to be orthogonal to Li and maintaining a differential angle Ψi which is an angle between the reference direction L and the center line CL. The probe P is moved while supplying an electric current,
The induced electromotive force generated in the pair of detection sensors 11a and 11b by the eddy current U generated by the reinforcing fiber 2 of the target cured layer 4i is converted into an electric signal, respectively.
The difference between the two converted electric signals (analog or digital signal) is calculated, and the swell angle φ of the fiber swell 8 portion of the reinforcing fiber 2 of the target cured layer 4i is calculated based on the difference value. ,
This is a method for detecting a fiber orientation defect of a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the waviness angle φ exceeds ± 2 °.

Claim 11 is a first example (FIGS. 1 (a) and 1 (b)) in which the first and second detection methods are further limited.
The fiber reinforcement according to any one of claims 7 to 10, wherein the converted two electric signals are converted into digital signals by A / D conversion after calculating and amplifying the difference as the analog signal. This is a method for detecting fiber orientation defects in a mold laminate.

Claim 12 is a second example (FIG. 1 (c)) in which the first and second detection methods are further limited.
The fiber-reinforced type according to any one of claims 7 to 10, wherein the two electric signals which are the basis of the difference calculation are digital signals obtained by converting the converted analog signal by A / D conversion. This is a method for detecting fiber orientation defects in a laminate.

13. The method for detecting a fiber orientation defect in a fiber-reinforced laminate according to claim 9 or 10.
As the arithmetic processing for calculating the portion exceeding the swell angle φ of ± 2 °, it is characterized in that filtering using the spatial Fourier transform and the Hough transform are used.

According to the probe P of the present invention, the exciting coil is the uniform exciting coil 10, and two or more detection sensors 11a to 11n are installed in a row in a direction orthogonal to the extending direction of the lead wire of the uniform exciting coil 10. Therefore, by selecting the arrangement direction of the uniform exciting coil 10, the cured layer 4i to be inspected can be freely selected from the laminated body 1. In addition, by using the uniform excitation coil 10, the difference in the detected values appears larger than that of the conventional probe Q, and a slight orientation deviation angle Δθ and swell angle with respect to the inspection orientation direction Li of the hardened layer 4i of the target. It has become possible to detect φ. As a result, a macroscopic misalignment having an orientation deviation angle Δθ exceeding ± 3 ° (greater than ± 3 °) and a local swell angle φ having a swell angle φ exceeding ± 2 ° (greater than ± 3 °), which was impossible in the past, It has become possible to detect target misalignment.

Further, when n detection sensors 11a to 11n are used, by selecting any two or more pairs thereof, inspection by two-dimensional measurement is possible only by scanning in one direction.
In the above, the "extending direction of the conducting wire" of the uniform exciting coil 10 is the direction in which the conducting wire extends on the exciting surface 10r of the uniform exciting coil 10 in which the detection sensors 11a to 11n are arranged, that is, is routed. It means that the detection sensors 11a to 11n are arranged in a row in the direction of the center line of the tubular uniform exciting coil 10.

In claim 2, if the detection sensors 11a to 11n are installed inside the uniform excitation coil 10, the uniform excitation coil 10 can be brought closer to the inspection surface 1a by the height of the detection sensors 11a to 11n, which is stronger. The eddy current U can be induced.

According to the probe P (FIG. 4) of claim 3, the eddy current of the eddy current U flowing through the laminate 1 under the uniform exciting coil 10 by partially changing the winding density of the uniform exciting coil 10 The density can be made uniform. As a result, not only the uniform exciting coil 10 can be shortened, but also the installation location of the detection sensors 11a to 11n is not limited, and the sensing accuracy can be improved.

According to the probe P (FIG. 5) of the fourth and fifth aspects, not only the eddy current density can be made uniform, but also the detection sensors 11a to 11n are installed in the intermediate portion 10c without the exciting coil. Because it can be formed or can be arranged inside, the exciting surface 10r of the winding start portion 10a and the winding end portion 10b can be brought closer to the inspection surface 1a, and a stronger eddy current U can be generated.
In this embodiment, in the detection sensors 11a to 11n, one or more detection sensors are arranged in a row at the winding start portion 10a and the winding end portion 10b, respectively.

The "inspection object" of claim 6 is a "pseudo-isotropic laminate 1" in the present invention, and includes, for example, curved members such as an airplane wing and a wind turbine for wind power generation. Since the exciting surface 10r of the uniform exciting coil 10 needs to have a constant separation distance from the inspection surface 1a of the inspection object, this exciting surface 10r is used for the inspection surface 1a having such a curved surface shape. It is preferably formed so as to match the curved surface-shaped inspection surface 1a.

According to claims 7 and 8, the orientation deviation angle Δθ of the hardened layer 4i, which is the target of the hardened layers 4a to 4n constituting the laminated body 1, can be detected with high accuracy (within ± 1 °). 9 and claim 10 can detect the waviness angle φ of the reinforcing fiber 2 of the hardened layer 4i, which is the target of the hardened layers 4a to 4n constituting the laminated body 1, with high accuracy (within ± 1 °).

(A) is the first example of the block diagram of the whole detection apparatus of this invention, (b) is the second example, and (c) is the third example. It is a perspective view of an example of the inspection state by the uniform excitation differential type probe of this invention. (A) is a perspective view of a first embodiment of the uniformly excited coil of the present invention, (b) is a schematic plan view thereof, (c) is a front view thereof, and (d) is a modified example thereof. (A) is a perspective view of a second embodiment of the uniformly excited coil of the present invention, (b) is a schematic plan view thereof, (c) is a front view thereof, and (d) is a modified example thereof. (A) is a perspective view of a third embodiment of the uniform exciting coil of the present invention, (b) is a schematic plan view thereof, and (c) and (d) are front views of an example in which the detection sensor is arranged in the center. (E) and (f) are front views of an example in which detection sensors are arranged at both ends. (A) is a front view of a uniform exciting coil using a large number of detection sensors, and (d) is a modified example thereof. It is an exploded perspective view which shows an example of the laminated state of the prepreg of a laminated body. (A) is a perspective view showing a macroscopic misalignment in the first cured layer of the laminated body, and (b) is a local misalignment and an enlarged view thereof. It is a drawing which shows the relationship between the fiber direction of a laminate, the differential angle of a probe, and the orientation direction of the reinforcing fiber of the hardened layer of an inspection target. (A) is a contour diagram of macroscopic misalignment, (b) is a spatial FFT processing diagram, (c) is a binarization diagram, and (d) is an analysis image by Hough transform. It is a measurement drawing of the waviness fiber length (local misalignment) of a reinforcing fiber. (A) The contour diagram of the fiber swell (local misalignment) of the reinforcing fiber, and (b) is the Hough transform diagram thereof. (A) is a plan view of the conventional probe and the probe of the present invention when the differential angle is 90 °, and (b) is a plan view when the differential angle is 0 °. (A) to (e) are diagrams showing the arrangement relationship of the detection sensor with respect to the fiber direction. It is an exploded view which shows the relationship of the eddy current of a probe and a hardened layer. (A) Macroscopic misalignment detectability evaluation table, (b) Local misalignment detectability evaluation table.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. First, the fiber-reinforced laminated body 1 to be inspected to which the present invention is applied (hereinafter, may be simply referred to as the laminated body 1) will be described. The laminated body 1 is obtained by laminating a plurality of B-stage prepregs 41a to 41n and heating them under pressure to cure them. The orientation direction of the reinforcing fibers 2 of the prepregs 41a to 41n and the number of laminated fibers are appropriately selected depending on the required strength of the members used.

The prepregs 41a to 41n used for laminating are a large number of conductive reinforcing fibers 2 aligned in one direction, or a conductive reinforcing fiber sheet woven such as a plain weave or a sardine weave, and the conductive reinforcing fibers. It is a semi-dried product (B stage) of a composite material with a matrix 5 (thermosetting resin, thermoplastic resin or ceramic) containing 2.

The orientation direction of these prepregs 41a to 41n is determined in advance at the design stage according to the purpose of use of the laminated body 1 which is a cured product, and at the manufacturing stage, the orientation direction of the conductive reinforcing fibers 2 indicated in the design. (This is indicated by the inspection orientation directions La to Ln. The subscripts in lower letters correspond to the cured layers 4a to 4n.) The layers are sequentially laminated. Therefore, the orientation direction of the conductive reinforcing fibers 2 of the cured layers 4a to 4n of the laminated body 1 which is the cured product is also known at the manufacturing stage. However, the misalignment as described above may occur in the laminated body 1 at the actual manufacturing stage (FIG. 8).

FIG. 7 shows an example in which four prepregs 41a to 41d are laminated for the sake of simplicity of explanation. The fiber directions of the prepregs 41a to 41d are indicated by thin line hatching. In the embodiment shown in the figure, the reinforcing fibers 2 of the prepregs 41a to 41d are laminated in a different orientation direction from, for example (0 °, 45 °, 90 °, −45 °), and the cured product of the reinforcing fibers 2 is laminated in FIG. The body 1 is a pseudo isotropic laminate 1. Although not shown, the method of laminating the cured layer 4 of the laminated body 1 is not limited to this.

The conductive reinforcing fibers 2 used include carbon fibers, boron fibers, ceramic fibers, fibrous materials in which highly conductive metals and graphite are uniformly dispersed in synthetic fibers, organic fibers, and inorganic fibers. Fibers coated with metal, fibrous materials whose surface is coated with a resin containing a conductive substance, metal fibers made by fiberizing metals such as stainless steel, copper, and aluminum, and carbon nanotubes. At least one of them (ie, one or a combination of two or more) is used.

The conductive reinforcing fibers 2 embedded in the matrix 5 which is the base material are electrically connected at the contact points with other adjacent conductive reinforcing fibers 2, and the eddy current U preferentially in the conductive reinforcing fibers 2. Is flowing. The non-conductive matrix 5 structurally functions in the same manner as a capacitor that repeatedly charges and discharges due to a high-frequency eddy current U, and is inferior to the conductive reinforcing fiber 2 in that the eddy current U flows.

For the matrix 5, a material suitable for the use of the laminated body 1 is used. In the case of thermosetting resin, for example, unsaturated polyester, epoxy resin, phenol resin, silicone resin and the like are used. The conductive reinforcing fiber 2 is included in the matrix 5 by the above-mentioned well-known method. Examples of the thermoplastic resin include polyamide (nylon) and a water-based adhesive in which vinyl acetate is dispersed in water. In addition, ceramic is also used.

The laminated body 1 as a product to be inspected is a flat plate in the embodiment shown in the figure, but there are various shapes such as a twisted one, a spherical one, and a curved one, and the inspection thereof is performed. The surface 1a is not limited to a flat surface. Therefore, the shape of the exciting surface 10r (the surface arranged to face the inspection surface 1a) of the probe P described below is made to match the shape of the laminated body 1, particularly the inspection surface 1a. As a result, a uniform separation distance H is maintained over the entire surface of the exciting surface 10r and the inspection surface 1a.

Further, since the laminated body 1 as a product is manufactured by being laminated based on the design as described above, the fiber directions of the cured layers 4a to 4n of the laminated body 1 are known in advance by the design. However, in actual production, since the prepregs 41a to 41n are laminated according to the design, it is considered that the orientation direction of the reinforcing fibers 2 of each of the cured layers 4a to 4n is a value approximately close to the design value. It is not always laminated without causing macroscopic misalignment (alignment deviation), and local misalignment (part of fiber swell 8) may occur during manufacturing (FIG. 8).

Therefore, it is necessary to determine the reference direction for the inspection of the laminated body 1 (this is referred to as the reference direction L), but the design is separated from the orientation direction of the reinforcing fibers 2 of the laminated body 1 to be inspected. The orientation direction of the reinforcing fibers is set to the reference direction L, and the cured layer of any one of the laminated bodies 1 to be inspected is selected, and the orientation direction of the reinforcing fibers is set to the reference direction L of the entire laminated body 1 to be inspected. In some cases. As described above, the orientation direction of the reinforcing fibers 2 of the cured layers 4a to 4n of the laminated body 1 to be inspected is substantially the same as the orientation direction of the reinforcing fibers in design. Therefore, for the sake of simplicity of explanation, the latter ( The orientation direction of the first cured layer 4a) is set to the reference direction L, and for the second and lower cured layers 4b to 4n, the design orientation direction is set to each cured layer 4b to 4n with respect to the reference direction L. Let the orientation directions Lb to Ln be 4n.

As the probe P, the one shown in FIG. 3 is taken as an example. In this probe P, two detection sensors 11a and 11b are arranged in a row in close proximity to each other along the center line of the exciting coil 10. The straight line connecting the centers of the detection sensors 11a and 11b is defined as the center line CL.
The probe P is arranged close to or in contact with the inspection surface 1a, and as will be described later, the center line CL thereof is parallel to or orthogonal to the orientation direction Li of the cured layer 4i to be the inspection target in consideration of the detection accuracy. Will be installed. The center line CL and the reference direction L intersect. The angle between the two is represented by the differential angle Ψi. The orientation direction Li of the cured layer 4i parallel to or orthogonal to the center line CL is the design orientation direction, intersects the reference direction L, and the angle formed by the two is the orientation angle θi of the cured layer 4i. Become.
When the actual orientation direction of the reinforcing fibers 2 of the cured layer 4i has a macroscopic misalignment with respect to the design orientation direction Li of the cured layer 4i, the intersection angle between the two is represented by an orientation deviation angle Δθ. Will be done.
The detection of macroscopic misalignment in the present invention is to detect an orientation deviation angle Δθ exceeding ± 3 ° in each of the cured layers 4a to 4n.

On the other hand, fiber waviness 8 which is a local misalignment may occur in any part of the cured layers 4a to 4n. FIG. 8B shows a case where fiber waviness 8 is generated in a part of the reinforcing fibers 2 of the cured layer 4a of the first layer. The enlarged view is a drawing substitute photograph of the actual fiber swell 8.
Assuming that fiber waviness 8 is generated in the hardened layer 4i of the inspection target, the size of the fiber waviness 8 is the tangent Si to the fiber waviness 8 and the actual reinforcing fiber 2 in the hardened layer 4i as shown in FIG. It is represented by the intersection angle (waviness angle φ).
The detection of local misalignment in the present invention is to detect a line having a waviness angle φ exceeding ± 2 ° in each cured layer 4b to 4n (or all cured layers 4a to 4n) and its length.
If the amount of macroscopic / local misalignment exceeds the threshold for good / bad judgment, it is judged as a defective product.

The probe P of the present invention is a uniform excitation differential type, and is basically composed of a uniform excitation coil 10 and two pairs of detection sensors 11a and 11b, and is housed in a casing (not shown). Other examples will be described later.
Hereinafter, the probe P shown in FIG. 3A is taken as a typical example. The uniformly excited differential probe P is in a fixed state in the embodiment of FIG. 1, but 2D moves relatively back and forth and left and right in order to collect data on the inspection range of the inspection surface 1a of the laminate 1 to be inspected. A scan is done. Therefore, the laminated body 1 is installed on the XY−θ stage 20. When the XY−θ stage 20 is not provided, the probe P may be attached to a driving body such as an articulated robot arm and moved.

The uniform exciting coil 10 shown in FIGS. 3 (a) to 3 (c) is the first embodiment thereof, and the exciting coil 10 has a conducting wire having an insulating coating formed on its surface at regular intervals (or in FIG. (Not shown, but tightly wound without gaps) It is wound in a spiral shape to form a flat rectangular parallelepiped. The exciting surface 10r of the exciting coil 10 arranged close to the inspection surface 1a of the laminated body 1 and parallel to some extent is made to match the shape of the inspection surface 1a of the laminated body 1.

Since the laminated body 1 has various shapes as described above, the uniform exciting coil 10 is not only wound in a rectangular parallelepiped shape as described above, but also has a cubic shape, a tubular shape, or the like according to the laminated body 1. Other shapes are also possible. The tubular shape includes a cylindrical shape, an elliptical tubular shape, a tubular triangular prism, a tubular polygonal prism, and the like. Then, as described above, in particular, the exciting surface 10r of the uniform exciting coil 10 is formed so as to match the inspection surface 1a of the laminated body 1. In other words, it is formed so that the separation distance H of the exciting surface 10r of the uniform exciting coil 10 from the inspection surface 1a of the laminated body 1 is maintained at substantially the same distance over the entire exciting surface 10r.

When a high-frequency current is passed through the uniform exciting coil 10 formed in this way, a fluctuating magnetic field is generated around the uniform exciting coil 10. When the laminated body 1 has pseudo isotropic property, a strongly directional eddy current mainly flows through the cured layer having the fiber orientation in the direction suitable for the fluctuating magnetic field among the hardened layers 4a to 4n constituting the laminated body 1. Almost no eddy current is generated in the other cured layers. Therefore, by selecting the arrangement direction (differential angle Ψ) of the uniform exciting coil 10, it is possible to select the cured layer in which the eddy current U flows with respect to the laminated body 1 having pseudo isotropic property.

Since the lead wire is wound evenly throughout the uniform exciting coil 10 of the first embodiment, the fluctuating magnetic fields around the winding start portions 10a and the winding end portions 10b at both ends gradually decrease toward the ends from the intermediate portion 10c. Therefore, the eddy current U caused by this also has a strong intermediate portion and gradually decreases at the winding start portion 10a and the winding end portion 10b.

As the detection sensors 11a to 11n, those having the same performance are used because the exciting coil 10 detects the induced electromotive force caused by the eddy current U generated on the inspection surface 1a. Examples of the detection sensors 11a to 11n used in the present invention include coils, Hall elements, MR elements, MI elements, and fluxgate sensors. Here, a coil is adopted as a typical example.
When the detection sensor 11 is a coil, in order to ensure the same performance, in the embodiment of the figure, it has the same circular shape (same diameter, same number of turns, same winding direction). In FIG. 3, the two detection sensors 11a and 11b are arranged so as to be close to each other with a slight gap in the center of the exciting surface 10r which is the lower surface of the uniform exciting coil 10. The arrangement direction is a direction orthogonal to the extension direction (winding direction) of the lower side 10d of the lead wire of the uniform exciting coil 10 (in other words, from the winding start side to the winding end side) in a row. There is.
Since the detection sensors 11a and 11b are used as a pair, there are two in this embodiment, but in the range where eddy currents of the same intensity and direction flow, three or more are installed, and any two of them are installed. It is also possible to obtain the differential of the signal of. It is also possible to use it in multiple pairs. This point will be described later.
The planar shapes of the detection sensors 11a and 11b in the figure are circular, but of course, they are not limited to this, and squares, polygons, and other appropriate shapes are adopted, but in order to make the sensing conditions the same as described above. It has the same shape.
In FIGS. 3A to 3C, the detection sensors 11a and 11b are arranged below the exciting surface 10r of the exciting coil 10. Therefore, the exciting surface 10r is separated from the inspection surface 1a of the laminated body 1 by the amount of the detection sensors 11a and 11b.

Next, a modified example of the first embodiment shown in FIG. 3D will be described. In this case, the detection sensors 11a and 11b are arranged inside the exciting coil 10, and are arranged close to and above the exciting surface 10r. As a result, the exciting surface 10r can be brought closer to the inspection surface 1a by the height of the detection sensors 11a and 11b as compared with the first embodiment, and a stronger eddy current U can be generated.
Whether the detection sensors 11a and 11b are installed above or below the exciting surface 10r also applies to the second embodiment and the following.

Next, a second embodiment of the uniform exciting coil 10 will be described (FIGS. 4A to 4D). In this uniform exciting coil 10, the winding start portion 10a and the winding end portion 10b of the lead wire are wound more densely than the intermediate portion 10c, and the two detection sensors 11a and 11b are installed in the intermediate portion 10c. Has been done. The winding method of the uniform exciting coil 10 is not limited to this, and even if the winding spacing is gradually reduced from the beginning to the center of the lead wire and the winding spacing is gradually reduced from the center to the winding end. Good. The two detection sensors 11a and 11b are arranged above or below the exciting surface 10r as described above.
As a result, the fluctuating magnetic fields around the winding start portion 10a and the winding end portion 10b appear stronger than in the case of the first embodiment, and the gradual decrease of the fluctuating magnetic fields at both end portions 10a and 10b is prevented, and as a result, the first The eddy current U generated by the uniform exciting coil 10 of the second embodiment is formed substantially uniformly over the entire width of the uniform exciting coil 10. Therefore, the uniform exciting coil 10 can be made smaller than that of the first embodiment.

A third embodiment of the uniform exciting coil 10 (FIGS. 5A to 5F) is composed of a winding start portion 10a and a winding end portion 10b without an intermediate portion 10c, and FIGS. 5 (c) and 5 (d). ), The two detection sensors 11a and 11b are installed in the center of the intermediate portion 10c, and the two detection sensors 11a and 11b are arranged above or below the exciting surface 10r.
In FIGS. 5 (e) and 5 (f), the two detection sensors 11a and 11b are arranged above or below the exciting surfaces 10r of the winding start portion 10a and the winding end portion 10b, respectively.
The winding directions of the lead wires of the winding start portion 10a and the winding end portion 10b are the same.
It is preferable that the lower surfaces of the detection sensors 11a and 11b (surfaces arranged close to (or in contact with) the inspection surface 1a of the laminated body 1) are installed so as to substantially coincide with the exciting surface 10a of the uniform exciting coil 10.
Since the uniform exciting coil 10 of the third embodiment has no intermediate portion 10c, the uniformity of the generated eddy current U is inferior to that of the first and second embodiments, but the exciting coil 10 and the inspection surface 1a Since the separation distance H is small, a stronger eddy current can be induced.

The detection sensors 11a and 11b are installed in a range (intermediate portion 10c) where the uniformity of the eddy current U to be detected is guaranteed.
The probe P of the present invention uses n detection coils 11a to 11n in a row, but in this embodiment, two exciting coils 10a at the start and end of winding having the same winding direction. Since it is composed of 10b, the magnetic field distribution has a concave shape at the center. Therefore, it is preferable to install two detection coils 11a and 11b in the central portion 10c.

6 (a) and 6 (b) are examples of the fourth embodiment in which three or more detection sensors 11a to 11n arranged in a row are arranged inside or outside the exciting coil 10. In the first to third embodiments, the number of detection sensors is two, but it is possible to have a plurality of detection sensors as in the fourth embodiment. In the case of FIG. 5, a plurality of winding start portions 10a and a plurality of winding end portions 10b can be installed. When there are a plurality of detection sensors 11a to 11n, any two are selected and the difference is taken. Further, if the probes P are arranged so that the detection sensors 11a to 11n are parallel to the traveling direction of the probes P and are moved in a horizontal row, a plurality of pairs are selected and an electric signal is recorded. Then, two-dimensional measurement becomes possible only by scanning in one direction.
The pair of detection sensors 11a to 11n to be selected are adjacent to each other, but it is also possible to select distant detection sensors depending on the detection conditions, for example, to detect a wide range of waviness.

Next, the configuration of the entire detection device of the present invention will be described (FIGS. 1A to 1C).
As shown in FIG. 6A, the detection device of the present invention performs differential processing of electrical signals (analog) from the detection sensors 11a and 11b with the differential amplifier 16 and as shown in FIG. , The electrical signal (analog) from the probe P is input to the lock-in amplifier 17 without using the differential amplifier 16, and the electrical signal (analog) is differentially processed and then amplified (or vice versa). Then, after amplifying the electric signal (analog), the difference is processed and output to the A / D converter 18), and as shown in the figure (c), the differential amplifier 16 and the lock-in amplifier. There are three ways in which the electrical signal (analog) from the probe P is A / D converted as it is and output to the computer 12 without using the 17 and the difference processing is performed by the computer 12. Hereinafter, the description will be centered on the figure (a), and the drawings (b) and (c) will mainly explain only the parts different from the figure (a).

The function synthesizer (or oscillator) 14 is a frequency oscillator capable of generating an arbitrary frequency, converts the AC current supplied from the power supply 30 into a predetermined high frequency current, and supplies the AC current to the uniformly excited coil 10. It should be noted that the high frequency pulse current may be supplied instead of the predetermined high frequency current. (In this specification, a high frequency pulse current is also included in the concept of the high frequency current in order to eliminate the complexity.) In the present embodiment, a high frequency current of 2 MHz or more is applied to the uniform exciting coil 10.

The differential amplifier 16 is a device that converts the induced electromotive force generated by the two detection sensors 11a and 11b into an analog electric signal and amplifies the difference (analog differential signal) of the analog electric signal.

The lock-in amplifier 17 uses the alternating current of an arbitrary frequency generated by the function synthesizer 14 as a reference signal, and among the differential signals amplified by the differential amplifier 16, the filtered component equal to the reference signal frequency. It is a device that amplifies and outputs only.

The A / D converter 18 is a device that converts an analog signal (differential signal) output from the lock-in amplifier 17 or the above two analog signals into a digital signal and outputs the digital signal to the computer 12.

The computer 12 performs a difference processing on the obtained digital signal (difference value) or the above two digital signals to perform a necessary image (creating a contour diagram), and calculates the orientation deviation angle Δθ and the waviness angle φ of the reinforcing fiber 2. At the same time, it controls the creation of an image analysis diagram and the movement control of the XY−θ stage 20, which will be described later.
In the computer 12, the spatial FFT image processing unit 12a, for example, the alignment angle calculation unit 12b for calculating the angle error (alignment deviation angle Δθ) with respect to the orientation direction Li of the target cured layer 4i, and the swell angle φ with respect to the fiber swell 8 are calculated. A swell angle calculation unit 12c, an output unit (not shown) for outputting the processing result and the calculation result to the monitor 13 and others are built-in.

The stage controller 19 is a device that controls (XY movement and rotational movement) the XY−θ stage 20 on which the laminated body 1 to be inspected is mounted.
Here, an example in which the XY−θ stage 20 is used for moving (forward / backward / horizontal rotation) of the laminated body 1 to be inspected is shown, but of course, the present invention is not limited to this, and as described above, for example. An articulated robot or the like may be used instead of the XY−θ stage 20 to realize the movement of the laminate 1. When the XY−θ stage 20 is used, an insulating sheet 21 is provided between the laminated body 1 and the XY−θ stage 20 for insulation.

The spatial FFT image processing unit 12a performs calculation of the difference value of the digital signal from the A / D converter 18 or by the computer 12 itself (FFT analysis of the spatial frequency of the difference value, and further acquisition of the FFT spectrum). If necessary, the filtering process and its waveform (contour diagram by inverse FFT and two-dimensional spatial frequency spectrum) are displayed on the monitor 13 (FIGS. 10A and 10B).

The orientation angle calculation unit 12b is a spatial FFT image processing unit 12a that binarizes a two-dimensional spatial frequency spectrum image (data), and further binarizes the binarized data, for example, by Hough transforming and inspecting the orientation direction (macroscopic view) of the reinforcing fiber 2. Misalignment: The alignment deviation angle Δθ) of the reinforcing fiber 2 of the cured layer 4i with respect to the design orientation direction Li for inspection is calculated with high accuracy. In the figure of FIG. 10D, the part of the emission line is the orientation direction of the detected reinforcing fibers of the target cured layer.

The swell angle calculation unit 12c performs a swell angle (local misalignment) of the reinforcing fiber 2 inspected by, for example, Hough transforming a contour diagram (data) filtered by the spatial FFT image processing unit 12a using an appropriate spatial frequency. Is calculated accurately. In the figure of FIG. 12B, the waviness angle φ is displayed together with the image of the detected reinforcing fiber of the target cured layer.

The monitor 13 is connected to the computer 12 and displays an image processed or calculated (a macroscopic and local misalignment detection image of the laminated body 1).

FIG. 1B shows a second example of this device configuration, in which the differential amplifier 16 is not provided and the analog electric signal caused by the induced electromotive force generated by the two detection sensors 11a and 11b is locked in the amplifier. It is sent to 17, and after differential processing by the lock-in amplifier 17, it is amplified (or amplified and then differential processed) and output to the A / D converter 18. The lock-in amplifier 17 is filtered by the reference signal frequency sent from the function synthesizer 14.

FIG. 1C shows a third example of the configuration of the present apparatus, which is caused by the induced electromotive force generated by the two detection sensors 11a and 11b without providing the differential amplifier 16 and the lock-in amplifier 17. The analog electric signal is sent to the A / D converter 18 as it is, converted into a digital signal, then sent to the computer 12, and the difference processing is performed by the computer 12. After that, only the filtered components that are also converted to digital signals and equal to the reference signal frequency sent to the computer are amplified.

Next, a method for detecting macroscopic / local misalignment by the probe P of the present invention will be described. As a typical example, the uniform excitation differential probe P to be used is as shown in FIG. 3 described above.
The laminated body 1 to be inspected is composed of cured layers 4a to 4n having different orientation directions, and the orientation direction of the reinforcing fibers 2 of the first layer is set as the reference direction L. It is assumed that the eddy current U due to the probe P is generated in all layers (Fig. 2).
In the following description, macroscopic and local misalignment of the reinforcing fiber 2 of the hardened layer 4i of the i-th layer is detected.

The probe P is installed on the upper surface of the first layer of the laminated body 1, that is, on the inspection surface 1a. Here, since the design orientation direction Li of the target layer i for inspection is known, the center line CL connecting the centers of the detection sensors 11a and 11b of the probe P is parallel to the inspection orientation direction Li. It is installed so as to (Fig. 9).
The angle formed by the center line CL and the reference direction L in this state is the differential angle Ψ, and the orientation angle θi formed by the design inspection orientation direction Li and the reference direction L in the target layer i. Is equal to (differential angle Ψ = orientation angle θi).
Although it is not known before the inspection, it is assumed that the orientation direction of the reinforcing fiber 2 of the target layer i has a misalignment of the orientation deviation angle Δθ with respect to the design orientation direction Li for inspection (FIG. 9). ).

The probe P two-dimensionally scans the inspection range of the laminated body 1 while maintaining the differential angle Ψ. The relative movement of the probe P with respect to the laminated body 1 is executed by the movement of the XY−θ stage 20 in the XY directions.

In the measurement, a high frequency current is passed through the uniform exciting coil 10 of the probe P (FIG. 15). As a result, the eddy current U preferentially flows through the hardened layer (i-2) having an orientation angle in the direction orthogonal to the differential angle Ψ in and around the uniform exciting coil 10 of the laminated body 1 (FIG. 15 (FIG. 15). e)).
However, since the fiber orientation angle θi of the hardened layer 4i of the target i-th layer is orthogonal to the winding direction of the uniform exciting coil 10, the eddy current U is difficult to flow, and the signal strength of the eddy current U is extremely high. It will be smaller (Fig. 15 (c)).
As shown in FIG. 15 (c'), when the orientation angle θi of the cured layer 4i of the target layer i forms a macroscopic misalignment with an orientation deviation angle Δθ, an eddy current U flows, which is essentially extremely high. The signal that should be small is strengthened, and it becomes possible to detect and evaluate the orientation deviation angle Δθ generated in the cured layer 4i.
Here, the angle of the eddy current U flowing in the hardened layers other than the hardened layer 4i of the target i-th layer is the design fiber orientation direction (L1 ... Li-1, Li + 1 ... Ln (not shown). )), Since it can be specified in advance by the spatial FFT processing unit 12a, it is possible to remove signals other than the cured layer 4i from the contour diagram by performing filtering for removing only signals other than the angle θi. By the above operation, the detection sensors 11a and 11b of the probe P preferentially record the eddy current U that flows when the orientation deviation angle Δθ exists in the hardened layer 4i that is the target.

As shown in FIG. 14A, in the target i-cured layer 4i, the detection coil 11a advancing at the head and the detection coil 11b at the tail advance in a row on the inspection orientation direction Li. When the orientation direction Li for inspection and the orientation direction of the reinforcing fiber 2 match (Ψ = θi, Δθ = 0 °), the front and rear detection coils 11a and 11b detect the eddy current U flowing through the same reinforcing fiber 2. Therefore, the difference is 0 (or the lowest value).

On the other hand, as can be seen from the above, in the uniform exciting coil 10, the generated eddy current U flows in one direction, so that the eddy current U flows only in the cured layer having the orientation direction corresponding to the eddy current U, and other than that. Aldy current U hardly flows in the hardened layer of. That is, in the conventional probe Q, an eddy current U'is generated around the entire circumference of the exciting coil 10', and the misaligned electric signal generated in the target cured layer can be extracted from the electric signal based on this eddy current U'. Although it is extremely difficult, it can be easily identified that only the electric signal of the target cured layer appears strongly or only the electric signal of the target cured layer appears weak as in the probe P of the present invention.
If the differential angle Ψ of the probe P of the present invention is adjusted so as to match or be orthogonal to the design fiber direction of the target cured layer, the signal from only the target cured layer is prioritized. Can be obtained.

Next, the accuracy of the uniformly excited differential probe P of the present invention will be described (FIGS. 13 and 16). The probe P of the present invention has an error (RMSE) with respect to a macroscopic misalignment in which the detectable angle (alignment deviation angle) is 7 ° at a differential angle Ψ = 90 ° (θi + 90 °) in the macroscopic misalignment. Is 1.3 ° and the differential angle Ψ = 0 ° (θi), the error (RMSE) is 0.8 ° for macroscopic misalignment where the detectable angle (alignment deviation angle) exceeds 3 °. Met.
From the above, using the probe P of the present invention with the “differential angle Ψ = θi” is optimal for detecting macroscopic misalignment, and for macroscopic misalignment exceeding the target detectable angle of 3 °. The detection error (RMSE) was 0.8 ° (about 1 °).

Next, regarding the detection of local misalignment (fiber swell 8), this swell part is a set of line segments on the contour diagram of the eddy current signal obtained by spatial Fourier filtering and Hough transform, and each line segment and The angle formed between the design orientation angles θi is the eddy angle φ.
The swell angle φ is detected as follows. The inspection area of the inspection surface 1a is two-dimensionally scanned to obtain a contour diagram of the eddy current signal caused by the eddy current U. The spatial frequency of this contour diagram is analyzed by the spatial FFT, and the high frequency spectrum is obtained from the low frequency. From this spectrum, noise is removed by a bandpass filter, and a spectrum with a frequency in the required range is extracted. As a result, a sharper contour diagram of the fiber swell 8 is obtained. After binarizing this contour diagram, for example, a Hough transform is used to obtain a line image corresponding to the reinforcing fibers (to be exact, a line image in which innumerable points are connected).
Then, among the points in the binarized image, those having a particularly strong peak in the part forming the line segment (the one having a high density of points of the points forming the line segment, or the number of points is high. Select an appropriate number (many ones).
Of the selected line segments, the threshold (detectable angle) is set as the target 2 °, and the swell angle φ between each line segment and the orientation angle θi exceeds 2 °, and the swell of each line segment is picked up. Error by comparing the angle with the true swell angle created by the tangent Si in the actual fiber direction that makes the center coordinates the same as each of the above line segments on the X-ray CT image that makes the scan range the same as the eddy current test. (RMSE) was calculated. At the differential angle Ψ = 90 ° (θi + 90 °), the error (RMSE) was 1.3 °.
That is, if the waviness angle φ is more than 2 °, it can be extracted with an error of 1.3 ° (about 1 °).

Unlike the probe Q of the conventional example, the probe P of the present invention filters the electric signals generated in other than the target hardened layer among the electric signals generated by the eddy currents generated in each hardened layer of the pseudo-isotropic laminate. Since only the electric signal from the target cured layer can be extracted, the macroscopic and local misalignment in the pseudo-isotropic laminate can be estimated more accurately.

CL: center line, H: separation distance, L: reference direction, La to Ln: inspection orientation direction, P: uniform excitation differential probe, Q: conventional circular excitation differential probe, Si: eddy current, U・ U': Eddy current, Ψ ・ Ψi: Differential angle, θ ・ θi: Orientation angle, Δθ: Orientation deviation angle, φ: Waviness angle, 1: (Conductive fiber reinforced) laminate, 1a: Inspection surface, 2: Conductive fiber 4.4a-4n: Hardened layer, 5: Matrix (resin part), 8: Fiber swell part, 10: Uniform exciting coil, 10': Circular exciting coil, 10a: Winding start part, 10b : End winding part, 10c: Intermediate part, 10d: Lower side, 10r: Exciting surface, 11a / 11b / 11a'・ 11b', 11n: Detection sensor, 12: Computer, 12a: Spatial FFT image processing unit, 12b: Orientation deviation Angle calculation unit, 12c: Eddy angle calculation unit, 13: Monitor, 14: Function synthesizer, 16: Differential amplifier, 17: Lock-in amplifier, 18: A / D converter, 19: Stage controller, 20: XY-θ stage, 21: Insulation sheet, 30: Power supply, 41a to 41n: Prepreg.

Claims (13)

A uniform exciting coil in which the lead wire is spirally wound to form a tubular shape,
Two or more arranged along one surface of the uniform exciting coil and installed in a row in a direction orthogonal to the extending direction of the conducting wire on the one surface, and any two of them. A uniform excitation differential probe characterized in that each is composed of a detection sensor used for detecting the difference in induced electromotive force due to excitation of a uniform excitation coil.
The uniform excitation differential probe according to claim 1, wherein the detection sensor is arranged outside or inside the uniform excitation coil.
The uniform excitation differential probe according to claim 1 or 2, wherein the winding start portion and the winding end portion of the lead wire of the uniform excitation coil are wound more densely than the intermediate portion thereof.
Claim 1 is characterized in that a uniform exciting coil is composed of a winding start portion and a winding end portion in the same winding direction without an intermediate portion, and two or more detection sensors are installed in the intermediate portion. Or the uniformly excited differential probe according to 2.
The uniform exciting coil is composed of a winding start portion and a winding end portion in the same winding direction without an intermediate portion, coincides with the winding start portion and the winding end portion, and detection sensors are installed inside or outside the winding end portion, respectively. The uniform excitation differential probe according to claim 1 or 2, wherein the probe is characterized in that.
According to any one of claims 1 to 5, the exciting surface of the uniform exciting coil arranged to face the inspection surface of the inspection object is formed according to the shape of the inspection surface of the inspection object. The uniform excitation differential probe described.
The fiber direction of any one of the cured layers of the fiber-reinforced laminate 1 in which a plurality of cured layers are laminated is set as the reference direction of the entire laminate.
The fiber direction determined in advance with respect to the reference direction is set as the inspection orientation direction of each of the remaining cured layers.
The line passing through the center of the selected pair of detection sensors of the uniformly excited differential probe according to any one of claims 1 to 5 is defined as the center line.
The center line is installed on the inspection surface side of the laminated body in accordance with the inspection orientation direction of the target cured layer, and the differential angle, which is the angle between the reference direction and the center line, is maintained. , The probe is moved while supplying a high frequency current to the exciting coil.
The induced electromotive force generated in the pair of detection sensors by the eddy current generated by the reinforcing fiber of the target hardened layer is converted into an electric signal, respectively.
The difference between the two converted electric signals is calculated, and based on the difference value, the orientation deviation angle of the reinforcing fiber of the target cured layer with respect to the inspection orientation direction is calculated.
A method for detecting a fiber orientation defect in a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the orientation deviation angle exceeds ± 3 °.
One reference direction is set for the fiber-reinforced laminate in which a plurality of cured layers are laminated.
The fiber direction of each cured layer, which is predetermined with respect to the reference direction, is set as the inspection orientation direction.
The center line passing through the center of the selected pair of detection sensors of the uniformly excited differential probe according to any one of claims 1 to 7 is aligned with the inspection orientation direction of the target cured layer of the laminated body. The probe is moved while supplying a high-frequency current to the exciting coil while being installed on the inspection surface side and maintaining a differential angle which is an angle between the reference direction and the center line.
The induced electromotive force generated in the pair of detection sensors by the eddy current generated by the reinforcing fiber of the target hardened layer is converted into an electric signal, respectively.
The difference between the two converted electric signals is calculated, and the orientation deviation angle of the reinforcing fiber of the target cured layer with respect to the inspection orientation direction is calculated based on the difference value.
A method for detecting a fiber orientation defect in a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the orientation deviation angle exceeds ± 3 °.
The fiber direction of any one of the cured layers of the fiber-reinforced laminate in which a plurality of cured layers are laminated is set as the overall reference direction of the laminate.
The fiber direction of each of the remaining cured layers, which is predetermined with respect to the reference direction, is set as the inspection orientation direction.
A laminated body in which the center line passing through the center of the selected pair of detection sensors of the uniformly excited differential probe according to any one of claims 1 to 7 is orthogonal to the inspection orientation direction of the target cured layer. The probe is moved while supplying a high-frequency current to the exciting coil while maintaining the differential angle which is the angle between the reference direction and the center line while being installed on the inspection surface side of the above.
The induced electromotive force generated in the pair of detection sensors by the eddy current generated by the reinforcing fiber of the target hardened layer is converted into an electric signal, respectively.
The difference between the two converted electric signals is calculated, and the swell angle of the fiber swell portion of the reinforcing fiber of the target cured layer is calculated based on the difference value.
A method for detecting a fiber orientation defect in a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the waviness angle exceeds ± 2 °.
One reference direction is set for the fiber-reinforced laminate in which a plurality of cured layers are laminated.
The fiber direction of each cured layer, which is predetermined with respect to the reference direction, is set as the inspection orientation direction.
A laminated body in which the center line passing through the center of the selected pair of detection sensors of the uniformly excited differential probe according to any one of claims 1 to 7 is orthogonal to the inspection orientation direction of the target cured layer. The probe is moved while supplying a high-frequency current to the exciting coil while maintaining the differential angle which is the angle between the reference direction and the center line while being installed on the inspection surface side of the above.
The induced electromotive force generated in the pair of detection sensors by the eddy current generated by the reinforcing fiber of the target hardened layer is converted into an electric signal, respectively.
The difference between the two converted electric signals is calculated, and the swell angle of the fiber swell portion of the reinforcing fiber of the target cured layer is calculated based on the difference value.
A method for detecting a fiber orientation defect in a fiber-reinforced laminated body, which comprises calculating and displaying a portion where the waviness angle exceeds ± 2 °.
The fiber reinforcement according to any one of claims 7 to 10, wherein the converted two electric signals are converted into digital signals by A / D conversion after calculating and amplifying the difference as the analog signal. A method for detecting fiber orientation defects in a mold laminate.
The fiber-reinforced laminate according to any one of claims 7 to 10, wherein the two electric signals that form the basis of the difference calculation are digital signals obtained by converting an analog signal by A / D conversion. Fiber orientation defect detection method.
The fiber of the fiber-reinforced laminated body according to claim 9 or 10, wherein filtering using a spatial Fourier transform and Hough transform are used as arithmetic processing for calculating a portion exceeding the waviness angle φ of ± 2 °. Orientation defect detection method.