By
applying an alternating magnetic field to the piece under examination, and measuring
the intensity of domain movement through a coil which causes an intermittent
electric sound (Barkhausen Noise) the noise level indicates the amount of stress
present.
The
intensity of the noise is influenced by different microstructural parameters
(size of the grain, hardness, direction of the fibres etc.) as well as, obviously,
the chemical composition of the material. Being sensitive to the variation of
these parameters, this method also allows detection of microstructurally discontinuous
areas generated, for example, by manufacturing faults, welding seams, and so
forth.
However,
separation of the effects of all these parameters is, at present, an open problem
and their simultaneous variation is perhaps the greatest limit to applying this
method. In particular, measuring the state of stress assumes importance when
it is possible to control the other parameters and therefore a calibration test
is needed for each material and microstructural state examined.
2. Correlation
of Barkhausen Noise-residual stress using diffractometric calibration
As mentioned previously,
the value of the Barkhausen Noise is influenced by both the presence of residual
stress and the microstructural parameters of the sample under examination. To
be able to identify areas suffering from "grinding burns" (with residual
stress close to the traction or even in the traction) using Barkhausen noise,
X-Ray Diffraction were applied
to evaluate both the state of stress and eventual microstructural alterations.
Correlating the results provided by the diffraction with the Barkhausen Noise
values allows us to create a "calibration curve" so we can attribute
residual stress values to the noise values. Obviously the Barkhausen/residual
stress conversion is valid only for the material being examined (type of steel,
heat treatment, roughness). Below we give the
calibration chart for manganese steel used for producing sprocket wheels. The
surfaces analysed had been ground.
Click on image to enlarge chart