Sizing limits of metal loss anomalies using tri-axial MFL measurements: A model study

A model study of the length and width sizing capabilities of the three components of the magnetic flux leakage (MFL) as used for inline inspections of pipelines is presented. The accuracy and limitations of the length and width sizing of metal loss anomalies are investigated using analytical results of the dipole model for the magnetic flux leakage of cuboidal anomalies. For anomalies (in particular metal loss) located at the internal pipe wall the accuracy of the length and width sizing is much better than for external anomalies. This is due to the fact that internal anomalies are located closer to the MFL sensors and thus are less smoothed in the measurements. In general the best results for the sizing are achieved using the radial component. The model results suggest that a further increase of both, the axial and circumferential resolution of current MFL tools can improve the sizing capabilities for internal anomalies, especially for internal pinholes. For external anomalies no significant improvement can be expected by increasing the axial sampling.

► The length and width sizing of MFL testing is investigated using the dipole model.
► The radial MFL component yields generally the best results.
► A better sampling rate and sensor spacing improves the sizing of internal anomalies.
► A higher axial sampling rate will not improve the sizing of external metal loss.