They get quoted against each other constantly, and they are not alternatives. One looks along the pipe and one looks through the cladding, and the damage each one cannot see is the thing that decides which you should be using.
Corrosion under insulation is the most expensive damage mechanism in a refinery, not because it is the most aggressive but because finding it means removing insulation, and removing insulation on a live plant costs more than the inspection. Both methods here exist to avoid that cost. They avoid it in opposite ways.
Guided wave ultrasonics (GWUT, or LRUT) clamps a ring of transducers around one bare band of pipe and sends a low-frequency wave along the pipe wall itself. The wall becomes the waveguide. The wave travels tens of metres in both directions and reflects off anything that changes the cross-sectional area — a weld, a support, a flange, and, if it is big enough, a patch of corrosion.
Pulsed eddy current (PEC) does something completely different. It drives a magnetic pulse down through the insulation and the cladding into the steel, then measures how fast the induced eddy currents decay. Thicker steel holds the current longer. The probe never touches the pipe and nothing is removed.
Guided wave inspects a length of pipe from one point. PEC inspects one point at a time, but anywhere you can reach. That single difference drives everything below.
It finds a change in cross-section. Typical practical sensitivity is somewhere around 5–9% loss of cross-sectional area at the ring — and that is area summed around the full circumference, not depth at a point.
Work through what that means. A 12-inch line with a 10 mm wall has roughly 9,900 mm² of metal in cross-section. Nine per cent of that is about 890 mm².
Guided wave cannot see pitting. Not "sees it poorly" — the physics of the measurement mean a pit does not register. If your damage mechanism produces pits rather than patches, guided wave is the wrong tool and no amount of operator skill fixes it.
It also cannot size anything. A call is "there is a change here of roughly this magnitude". Turning that into a wall thickness needs a second inspection at that location.
And it attenuates. Range falls off sharply with anything that drains energy from the wave: heavy general corrosion, bitumen or dense coating, buried sections, and every branch, bend and support along the way. A quoted "50 metres each way" is an unobstructed straight run in good condition. On a real rack you may get fifteen.
PEC reads average wall thickness over its footprint, and that footprint is large — typically tens of millimetres across, and it grows with the insulation standoff. The thicker the insulation, the wider the area being averaged.
Do the same arithmetic. A probe averaging over an 80 mm circle contains about 5,000 mm² of surface. A 15 mm pit through 80% of a 10 mm wall removes metal from about 3.5% of that area. The averaged reading drops by roughly 3%, which is inside the noise. PEC misses isolated pitting too — for a different reason, but with the same outcome.
PEC also gives you a relative value rather than an absolute thickness. It compares each reading against a reference established on a section believed to be sound. If your reference is already corroded, everything reads normal. This is the most common way a PEC survey goes quietly wrong.
| Guided wave | Pulsed eddy current | |
|---|---|---|
| Coverage per set-up | Tens of metres of pipe | One footprint, tens of mm |
| Insulation removal | One bare band per test point | None at all |
| Detects broad wall loss | Yes, above ~5–9% CSA | Yes, averaged |
| Detects isolated pitting | No | No |
| Gives a thickness | No | Relative only |
| Works through cladding | N/A — needs bare band | Yes, and through concrete and marine growth |
| Beaten by | Bends, branches, supports, heavy coating, burial | Thick insulation, geometry changes, bad reference |
| Best at | Long inaccessible runs — road crossings, sleeved sections, racks | Vessels, large-bore lines, awkward geometry, anywhere you can physically reach |
A buried road crossing. A line in a sleeve. A run across a rack thirty metres up. Anywhere the cost is not the inspection but reaching the pipe. Guided wave turns one accessible point into coverage of a length you could not otherwise touch, and that is a genuinely large saving.
Vessel walls, large-bore lines at grade, insulated equipment in a live unit. PEC lets you build a grid of readings over an area without a single square metre of insulation coming off, and it tolerates cladding, weather sheeting, concrete fireproofing and marine growth.
This is the important one. If the mechanism you are chasing produces pitting — chloride pitting under wet insulation on stainless, or microbiological attack — both of these methods will hand you a clean report on a badly damaged line. You need insulation off and either ultrasonic corrosion mapping or, on small bore, profile radiography.
A screening survey comes back clean, the line is signed off for another cycle, and it leaks eighteen months later from a pit that was there the whole time. Nobody did anything wrong except choose a method whose blind spot matched the damage. This is why "what does it miss" belongs in the inspection plan next to "what does it find".
They use these as screening tools and never as the answer. The sequence that works:
Steps 2 and 3 are not alternatives. Screening tells you where to look; it does not tell you whether the line is fit to run.
The guide lists every method that detects CUI, with what each one misses, and the mechanism card that explains why insulated lines corrode in the first place.
Open the detection methodsWho writes this. A mechanical engineer with twelve years in oil and gas — in-line inspection, fired heater and furnace inspection, and pipeline integrity. What is here comes from the published standards and from what those years in the field actually looked like. It is not written by an API-certified inspector.
This is not an assessment. Nothing on this site may be used to justify a decision about real equipment. Assessing plant requires the current editions of the applicable codes, data from a licensed source, and a competent engineer who signs for the answer. · Integrity Field Guide