"Electromagnetic Inspection"
-«
This invention relates to the inspection of materials by electromagnetic induction. Such inspection may be for the purpose of detecting defects (e.g. cracks) in a test piece which should be homogeneous,
5 or for other purposes such as identifying materials which are buried within a surface accessible to an inspection probe.
There are a variety of known inspection methods using electromagnetic induction. A common technique
10 is eddy current inspection, which relies on a test probe inducing eddy currents in a conductive sample, and then detecting changes in the induced currents when the probe is traversed over a flaw. These known techniques, however, suffer from serious limitations
15 in practical use.
An object of the present invention is to provide an improved inspection apparatus. A further object is to provide an inspection apparatus which is suitable fc-r interfacing with a computer for automated monitoring
20 and analysis.
The present invention provides apparatus for inspecting materials, including a probe comprising a winding formed adjacent one end of a member of high magnetic susceptibility (preferably a ferrite
25 rod) , a capacitance connected in parallel with said winding to form an LC resonant circuit, and a signal driver arranged to apply square-wave signals to said resonant circuit.
Preferably, the apparatus further includes means 30 connected to monitor the voltage across said resonant circuit and adapted to identify the amplitude and/or position of harmonic distortion in the waveform of said voltage.
In one form, the apparatus comprises a plurality of probes arranged in a matrix, and switching means are provided for connecting the signal driver to the probes in a desired order. Embodiments of the invention will now be described, by way of example only, with reference to the drawings, in which:-
Fig. 1 is a diagrammatic side view of a probe with associated circuitry, forming one embodiment of the invention;
Fig. 2 illustrates waveforms obtained in the apparatus of Fig. 1;
Fig. 3 shows a further embodiment;
Fig. 4 is a schematic plan view of a probe matrix used in another embodiment;
Fig. 5 is a cross-section on the line 5-5 of Fig. 4; and
Fig. 6 is a schematic plan view of another matrix.
Referring to Fig. 1, the apparatus of the invention includes a probe 10 comprising a coil 12 wound around one end of a ferrite rod 14. A capacitor 16 is con¬ nected in parallel with the coil 12 to form an induc¬ tance-capacitance (LC) resonant circuit. A signal driver 18 having a suitable output impedance is arranged to apply square-wave signals to the probe 10, and a monitor 19 is connected to monitor the wave¬ form across the LC circuit as will be discussed in greater detail below.
The coil 12 and capacitor 16 are selected to give a resonant frequency suitable for the material to be investigated. With the probe 10 in air, the frequency of the driving square-wave signal is adjusted to give a substantially sinusoidal waveform across the circuit. If the probe 10 is then brought into proxi- mity with a metallic test piece, indicated at 20,
the magnetic field produced by the probe 10 is coupled to the test piece 20 thus altering the load on the coil 12. The frequency is adjusted until the waveform across the LC circuit is again substantially sinusoidal. The coupling with the test piece 20 has the effect of enhancing third-harmonic distortion of the waveform across the LC circuit, producing distinct lobes 22 on the waveform 24, as seen in Fig. 2. If the probe 10 is now moved across the surface of the test piece 20, 0 the waveform 24 remains the same so long as the material of the test piece is homogeneous, but if some non- homogeneity occurs (e.g. a crack or an inclusion) the third-harmonic lobes change significantly in both amp¬ litude and pQsition, as indicated at 22a. 5 The monitor 19 may be any suitable means for detecting changes in the lobes 22. In the simplest case,, a cathode-ray tube could be used. Generally, however, it will be preferred to provide some form of automatic means to detect changes and initiate 0 an alarm or make a record. This could be done for example by digitising the waveform 24 and subjecting the data to computer analysis to determine the third- harmonic amplitude and its timing in relation to the driving square wave, or by filtering out the third- 5 harmonic and measuring its amplitude and/or timing. The time shift of the third-harmonic is three times that of the fundamental. Similarly that of the fifth harmonic is five times. Monitoring of odd harmonics thus gives a Vernier effect which magnifies
30 time domain changes caused by non-uniformities in the test piece. However, it will be understood that the amplitude of the harmonic distortion decreases substantially with each step up the odd harmonic series; thus the third or possibly fifth harmonic is most
35. suitable to monitor.
It would also be possible to provide automatic control of the signal driver repetition rate by feed¬ back from the monitor.
With a single, movable probe of the type discussed above, it is convenient to incorporate the LC circuit, signal generator and monitoring means in a hand-held unit.
The invention can also be applied to a matrix arrangement of probes positioned adjacent a test piece, with the driving signal being scanned from probe to probe, rather than moving a single probe. In this connection, attention is directed to our copending application PCT/GB85/00301, in which a somewhat similar matrix approach is disclosed. Referring to Fig. 3, a matrix of six probes 10 is shown. Signal generator 18 provides driving square wave signals on a bus 30. Each probe can be selectively coupled to the bus 30 by gates 32. The waveform produced across any given probe can be coupled by gates 34 to a second bus 36 and thence to a monitor and control circuit 38, which is suitably provided by a microprocessor. (Return conductors are omitted in Fig. 3 for clarity.) In use, the gates 32,34 for a given probe 10 are operated, the LC circuit driven to resonance, the microprocessor 38 checks that resonance has been achieved, analyses the third- harmonic component, and then controls the gates to activate a subsequent probe. It will be appreciated that Fig. 3 shows a very small matrix by way of example, and that in practice a much larger matrix would be used, in which case other switching arrange¬ ments might be suitable.
The matrix may be addressed in a sequential manner similar to a raster scan. It is also possible by suitable programming of the microprocessor to identify
the location of a possible anomaly and to investigate that area further by scanning in other directions across it.
With a large matrix, more than one probe may be driven at any given time provided there is sufficient separation between the probes to avoid interference.
Referring now to Figs. 4 and 5, a preferred form of matrix is shown. The matrix is formed on a flexible plastics substrate 40. Each sensor comprises a spiral coil 42 having a ferrite bead 44 superimposed on it. The coils 42 are formed by printing. It is necessary for the coils 42 to have a high conductivity and it is therefore preferred that they are of gold or silver. The centre of each spiral 42 is connected in circuit by means of a lead 46 passing through a central bore 48 in the ferrite bead 44. The conductors suitably are also printed on the substrate 40 as indi¬ cated at 50. The capacitor for each sensor may be a discrete component, as at 52, or may be formed on the substrate as at 54 by sequential printing or deposition of dielectric and conductive layers. Alternatively (not shown) the capacitor could be formed integrally on the ferrite bead.
The use of a flexible substrate is preferred, since this can be attached to a shaped former suitable for investigating for example welds in angled joints. However a rigid substrate would be suitable for many applications.
Turning to Fig. 6, there is shown schematically a matrix array of six probes 10 mounted on a carrier
60. The carrier 60 in turn is slidably housed within a frame 62 for repetitive movement around the path ABCD by suitable drive means (not shown) . The move¬ ment in each direction should be at least equal to the maximum matrix dimension, in this example of diagonal E,
which ensures that every part of the specimen under the frame 62 is inspected by at least three probes. This principle may be applied to geometries other than that shown; for example the matrix may be triangular, and the frame could be circular with the matrix carrier describing an orbital path. This type of arrangement is the subject of and more fully described in our British Patent Application entitled "Scanning Mechanism for Inspection Apparatus" filed 15th October, 1985.
The invention may be used not only to locate defects in a uniform material but also to distinguish between distinct materials. One application of this relates to tags of*the type described in our copending application PCT/GB85/00301, in which inserts of (for example) steel and ferrite form a coded array. The present apparatus gives separate, distinct harmonic responses to these materials and it is thus possible to regard one material as digital "0", the other as- digital "1", and to distinguish both from the mere absence of code.