This is the opening module, and it answers the question people ask: where is guided wave testing actually used? The answer is the part of pipeline you cannot reach.
A pipe under insulation. A high temperature pipe. A pipe in high-height. A pipe through a dike wall. A pipe through concrete penetration. A pipe under a road, under water, under the ground. In every one of those, the section that most needs inspecting is the section no conventional method can measure without excavation.
So the usual framing is backwards. Guided wave testing does not inspect from where the damage is. It inspects from where you can stand. The probe goes on the accessible section of pipe, and the wave travels into the inaccessible section and comes back.
That also tells you when not to use it. If the pipe is fully accessible, ultrasonic thickness measurement will give you a better answer. This technique earns its place where access is the problem.
The second half of the module covers long-term monitoring, and there is one idea in it that I think is under-appreciated in our industry.
In a single inspection, a defect must stand above the background noise of that record — roughly 2-to-1 before anyone can call it. That noise is not electronic. It is the pipe itself: surface condition, coating, weld caps, mode conversion.
In monitoring, you are not comparing a signal against noise. You are comparing today's record against the same probe, in the same position, last year. The welds, the supports, the coating all produce the same signal they produced before, and they subtract out. What remains is the change.
That turns a detection problem into a change-detection problem, which is far easier. A permanent sensor can follow a defect growing well below anything you could call from a single inspection.
Wave velocity depends on the shear modulus of steel, which falls as temperature rises — roughly 1% per 55 to 60 degrees C. Measure a line in January and again in July at a different process temperature and the signals shift in apparent distance, though nothing on the pipe has changed. Any monitoring system has to correct for this before two records can be subtracted. GIM does it by measuring the velocity and amplitude change against known reference signals in the same record, and calibrating both out before the subtraction.
Underneath all of it is the standard this whole series is held to, stated in the first two minutes: everything we produce is either evidence or an assertion. The number in a report the client cannot check is an assertion. The number supported by data the client can look at is evidence.