No attempt will be made here to describe all the intricacies of eddy-current NDT. Only a basic theoretical description is given. If you need more information on this subject there are lots of books available, and we strongly recommend the ASNT book Fundamentals of Eddy Current Testing, by Donald Hagemaier.
The principle of eddy-current testing is based on the capability of an alternating current in a coil to induce currents, eddy-currents, in a nearby test piece. The alternating current in the coil creates an alternating magnetic field which, when the coil is moved sufficiently close, penetrates into a test piece. If the work-piece is a conductor, i.e. has non-zero conductivity, the alternating magnetic field induces currents, eddy-currents, in the work-piece. The eddy-current itself induces a magnetic field. This field opposes the coil's magnetic field.
The opposing magnetic field created by the eddy-current in the work-piece changes the inductance and resistance of the coil. These two quantities make up a vector quantity called impedance. It is this small change in impedance, relative to the free-space inductance of the coil, the inductance when the coil is in air, which is displayed on the cathode or liquid crystal screen of eddy-current instruments. Materials with different conductivities and permeabilities will give different deflections. This is the basis of material characterization using eddy-current testing.
If the coil is placed on the work-piece and the deflected impedance signal noted, the signal will change as the coil is moved away, lifted off, the surface. This change in signal is the lift-off signal. The phase of this signal is important, because most positioning errors, like probe tilting or surface irregularities, cause noise in-phase with this signal.
If the eddy-current in the work-piece encounters a defect, the eddy-current must flow around or underneath it, if possible. This in turn changes the induced magnetic field; therefore changing the impedance of the coil. This deflected signal is what is used for defect location, characterization and sizing in eddy-current testing. The strength and phase of this signal are influence by many factors like test frequency, coil windings, defect dimensions, instrument sensitivity and noise.
The coil described above can be thought of as a probe. Indeed many eddy-current
probes are just simple windings encased in some non-conducting housing. Other
probes have ferrite cores made up of a non-conducting highly permeable
material. The introduction of this highly permeable material near the coil can
strengthen the magnetic field and hence the signal, but it does not alter the
discussion above. Eddy-Current modelling is an attempt to describe the system
behaviour with a simple set of rules. The accuracy of the model is dependent
on how well these rules describe the system. For example, if the system to be
studied is the population of rabbits and foxes in a closed ecological system,
the rules could be: foxes eat rabbits, rabbits don't eat foxes, rabbits eat
grass, grass dies if there are too many rabbits, foxes die if there are too
few rabbits. With these simple rules, an attempt can be made to model the population
changes inside this closed ecological system. This model as stated is not quantitative,
it just sets own rules or assumptions about the behaviour of the system.
It is the physical model.
|
|