Thomas W. Krause

Measurement of residual stress in steel using the magnetic Barkhausen noise technique

Surface residual stresses on a structural beam steel sample were evaluated using a non-destructive technique based on the measurement of surface magnetic Barkhausen noise (MBN). MBN measurements were performed using a high resolution probe consisting of a small magnetic read head mounted between the poles of a ferrite U-core magnet. Applied stress magnitudes were correlated to MBN energy levels for calibration purposes. MBN energy measurements were done at various locations on the steel sample. The magnitude of the residual stress component along the sample beam axis was evaluated across the width of the beam using these calibration curves. The range of sensitivity of the MBN signal to stress as defined by the calibration curves was limited by the direction of easy magnetization. The effect of stress on the MBN signal was interpreted in terms of the active 180° domain wall population. The validity of the residual stress results was confirmed experimentally using conventional methods: cutting and sectioning and hole drilling. Possible sources of residual stresses and their distribution on the structural steel specimen are discussed.

Investigation of the stress‐dependent magnetic easy axis in steel using magnetic Barkhausen noise

Angular‐dependent magnetic Barkhausen noise (MBN) measurements were performed on a pipeline steel sample for various values of applied uniaxial stress at three angles with respect to the sample’s zero stress magnetic easy axis direction. It was observed that the response of the MBN signal to stress was dependent upon the direction of the stress with respect to the zero stress easy axis. The stress response of the MBN signal was greatest for (i) tensile stresses oriented perpendicular to the zero stress easy axis direction and (ii) compressive stresses applied parallel to the easy axis direction. The modification of the MBN signal under an applied stress was attributed primarily to a change in the 180° domain wall population in the material investigated. Results were described by a model that considered regions of locally correlated domain behavior, termed ‘‘interaction regions,’’ that were typically the size of grains within the steel material. A basic result of the model was the stress required to modify...

CHARACTERIZATION OF THE MAGNETIC EASY AXIS IN PIPELINE STEEL USING MAGNETIC BARKHAUSEN NOISE

The angular dependence of magnetic Barkhausen noise (MBN) on eight surfaces through the thickness of a 2% Mn steel pipeline sample was investigated. The MBN signal was analyzed by integrating the square of the MBN voltage signal with respect to the time axis. The resulting value, referred to as the MBN energy signal, was modeled by considering the irreversible motion of 180° domain walls, under the influence of an oriented magnetic field. An expression for the angular dependence of the MBN energy signal was derived and was given by energy=αu2009cos2u2009θ+β, where α and β are the fitting parameters and θ is the angle between the maximum MBN signal and the applied sweep field. The α parameter was associated with the irreversible motion of 180° walls that contributed to the net macroscopic easy axis near the surface of the sample, while the β parameter was associated with the isotropic background MBN signal. The energy equation could be used to fit the data for all sweep field amplitudes in which the MBN spectrum w...

Investigations of magnetic flux leakage and magnetic Barkhausen noise signals from pipeline steel

Magnetic flux leakage (MFL) is used for in-line inspection of underground oil and gas pipelines. MFL signals are sensitive to the line pressure which should be taken into account when estimating the sizes of defects. MFL signals have been studied from electrochemically milled pits with 50% penetration in pipeline steel under the influence of different axial or circumferential stresses and a range of magnetic flux densities. Magnetic Barkhausen noise (MBN) measurements have been used to study the changes in the direction of the magnetic easy axis of the pipe wall. These help one to understand the stress-dependent MFL results. MBN measurements have also been used to estimate the directional magnetic anisotropies of the steels.

Characterization of texture and residual stress in a section of 610 mm pipeline steel

Abstract Gas pipelines are inspected for defects such as corrosion. The most commonly used nondestructive inspection tool uses the magnetic flux leakage (MFL) technique. The MFL signals depend on the magnetic behaviour of the pipe, which is sensitive to its microstructure and crystallographic texture as well as both residual and applied stresses. Here a section of commercial X70 pipeline is characterized using microstructural examination, X-ray diffraction (to determine crystallographic texture) and neutron diffraction (for residual stress measurement). The results correlate well with the manufacturing steps used for this type of pipe. Magnetic characterization is also performed using magnetic Barkhausen noise measurements, which reflect the magnetic anisotropy in the pipe and thus the MFL signal. These results do not correlate simply with crystallographic texture and residual strain results, but this is not unexpected given the complex nature of the material and its stress state.

Investigation of the magnetic field and stress dependence of 180° domain wall motion in pipeline steel using magnetic Barkhausen noise

Abstract Magnetic Barkhausen noise (MBN) measurements were performed on 2% Mn pipeline steel samples as a function of magnetic field (H) and as a function of orientation of an applied sweep field for various tensile and compressive stress values. For the magnetic field dependent measurements, the H field and stress were applied along the same axis, which was aligned within 10° of the samples magnetic easy axis. Tensile stresses between 210 MPa and 0 and compressive stresses between 0 and -200 MPa were applied. The MBN signal was observed to initially increase with H followed by a decrease and eventual suppression of the MBN signal at higher H values. The application of tensile stress increased the total MBN signal and increased the field at which suppression occurred, while the application of compressive stress decreased the total signal with a corresponding decrease in the suppression field value. The stress and low-field behaviour of the MBN signal was correlated with the irreversible differential permeability (μIDP) measured in a similar 2% Mn pipeline steel sample. The behaviour of the MBN signal as a function of H and stress was attributed to the effect of these parameters on the relative population and/or size of 180° domain walls, that is, the total 180° domain wall area, within the sample.

Investigation of strain dependent magnetic Barkhausen noise in steel

Angular dependent magnetic Barkhausen noise (MBN) measurements were performed on three pipeline steel samples under uniaxial tensile stress up to 260 MPa. An expression for the angular dependent strain given by /spl delta/L/L/spl ap//spl Delta/L/L(1+/spl nu/)sin/sup 2//spl theta/, where /spl theta/ is the angle of the strain component with respect to the maximum strain /spl Delta/L/L, was compared with the results. It was observed that the MBN signal was greater for the magnetic sweep field aligned along directions of greater tensile strain. Results were interpreted in terms of the relative increase of the active 180/spl deg/ domain wall population in the direction of the applied magnetic field and in the sense of increasing tensile strain.

Origin of a magnetic easy axis in pipeline steel

Oil and gas pipelines are generally magnetically anisotropic, with a magnetic easy axis in the pipe axial direction. This is of interest because magnetic flux leakage tools are commonly used for the detection and sizing of defects. In the present study we investigate the origin of this magnetic easy axis, using an angular magnetic Barkhausen noise technique to characterize the magnetic anisotropy. The texture, microstructure, and residual stress are examined as possible causes of the easy axis, using x-ray pole figure analysis and microstructural examination along with high and low temperature annealing treatments. Our results indicate that plastic deformation and residual stress are responsible for the magnetic easy axis, since an elimination of the residual stresses through low temperature “stress relief” heat treatment produces a magnetically isotropic structure without altering the texture or microstructure. X-ray pole figure analysis supports the conclusion that magnetic anisotropy is not related to texture in these materials. We conclude that the axial magnetic easy axis is due to a compressive residual hoop stress resulting from the cold bending and cold expansion of the pipe during processing.

Detection of stress concentrations around a defect by magnetic Barkhausen noise measurements

The stress distribution around a 50% blind‐hole pit in a steel pipe with a 9 mm wall has been studied using high‐resolution magnetic Barkhausen noise (MBN) measurements. A magnetic disk read‐head is used as the pick up coil in the MBN probe. The study shows a stress concentration factor of ∼2 at the defect edge perpendicular to the direction of applied stress and ∼−0.6 at the edge parallel to the same. The experimental results are consistent with the analytical solutions obtained by the Airy’s stress function approach.

Correlation of magnetic Barkhausen noise with core loss in oriented 3% Si–Fe steel laminates

The angular dependence of surface magnetic Barkhausen noise (MBN) with flux densities up to 1.7 T on eight different samples of oriented 3% Si–Fe steel laminate was investigated. On two of the samples encircle MBN measurements were performed. MBN energy values were obtained by integrating the square of the MBN voltage signal with respect to time. The angular variation of the MBN energy signal was modeled by considering anisotropic internal fields that moderate 180° domain wall motion and the local eddy current field interactions between spatially correlated Barkhausen events. The eddy current field interactions arising between 180° domain walls that lie along the sample’s primary easy axis direction were represented by a parameter, α′. α′ had a cosu2009θ angular dependence where θ is the angle of the applied sweep field with respect to the sample rolling direction. The angular averaged MBN energy signal, 〈Energy〉, was evaluated and compared with core losses measured using the standard Epstein technique at 1.5...