There have been a few instances where a defect (also known as a pipeline feature) has been located erroneously because pipeline maintainers have not been able to ascertain their bearings along the pipeline. As a result, the wrong pipe length has been repaired and the defect has remained present, making it a site for potential loss of containment. As well as being hugely damaging to the environment, loss of containment can cost the responsible company millions of dollars in remediation.

An in-line inspection (ILI) report provides the following details about a pipeline feature:

  • Length and number of pipe in which the feature exists, and length of pipes in the vicinity;
  • Internal or external dimension of the feature – length, width and maximum depth of the feature (box), along with its B-scan and C-scan;
  • Orientation of the feature, also known as the ‘o’clock position’ of the feature;
  • Distance of the feature (box) from upstream weld; and,
  • Distance of the feature (box) from downstream weld.

Once these details have been logged, the pipeline is then geographically located using an XYZ ILI tool, which maps pipe circumferential welds against their ‘northing and easting’ log distance. However, these readings have errors in them and therefore take one to the general vicinity of the feature but do not confirm the exactness of its location.

Article continues below…
Advertisement

Recognising the need for a foolproof method of confirming the location of pipeline features, we came up with the five-point check technique. It is a set of dimensions that compares report measurements against actual measurements, thereby increasing confidence in locating the pipeline feature. The dimensions are:

  • Distance of feature from upstream weld (report vs actual);
  • Distance of feature from downstream weld (report vs actual);
  • Feature dimensions, i.e. length, width and depth (report vs actual);
  • Feature orientation, i.e. o’clock position (report vs actual); and,
  • Location of the pipe with respect to pipe lengths in the vicinity.

The fifth point would only need to be confirmed when the first four points do not offer confidence.

It is ideal if the pipe length on which the feature occurs is different in some way, for example, it has a spool length that is different from the others, or has a valve, bend or a flange. In the absence of such a unique feature, one should take advantage of the trend of subtle differences in the pipe lengths.