Short answer: fired heater tubes are inspected for creep, overheating and metal loss on both the fire side and the process side. A sound programme combines on-stream infrared thermography of tube metal temperature, a shutdown visual examination of the radiant and convection sections, ultrasonic thickness and diameter measurement against a baseline, and, for serpentine coils that can be pigged, an in-line ultrasonic scan of the whole coil. API RP 573 sets out the inspection practice; API Standard 530 is where tube thickness and creep allowance come from.
\n\nMost pressure equipment is inspected for one question: how much wall is left. A fired heater tube adds a second question that thickness cannot answer: how much creep life has the tube used. Radiant tubes run in the creep range of their alloy, so a tube that ran too hot for too long can bulge and rupture while its wall is still close to nominal. The inspection plan has to collect evidence of temperature history and deformation as well as thickness.
\nThe tube is also attacked from both sides at once. Inside, the process stream can thin the bore through sulfidation or naphthenic acid in crude and vacuum service, or lay down coke that insulates the wall and pushes metal temperature up. Outside, the flue gas oxidises and scales the surface, and in the convection section the coolest rows can suffer flue-gas dew point corrosion. A single method rarely sees all of this.
\n\nAPI RP 571 describes each of these in detail; the short list for a refinery heater is:
\nWhich of these apply depends on the heater's service, fuel, alloy and operating history, so the damage mechanism review comes before the method selection, not after it.
\n\n| Method | What it tells you | Limits to plan around |
|---|---|---|
| Infrared thermography, on stream | Tube metal temperature across the radiant box, hot spots, flame impingement, coke build-up trends | Readings need correction for emissivity, scale and reflected flame radiation; only tubes visible from the sight doors are covered |
| Visual examination, at shutdown | Bulging, sagging, bowing, scale, hot-spot discoloration, damaged hangers and guides, refractory condition | Shows symptoms, not wall thickness or remaining life |
| Ultrasonic thickness on prepared spots | Remaining wall at the measured points, trended against previous outages | Scale must be removed for reliable readings; spot coverage misses localised loss between points |
| Diameter or circumference measurement | Creep swell compared with the as-built or previous diameter | Needs a trusted baseline; meaningful only on tubes in creep service |
| In-line ultrasonic pigging of the coil | Wall thickness and geometry, including bulging and ovality, along the whole coil | Run by specialist vendors; the coil must be piggable through its return bends or plug headers |
| Hardness and alloy verification (PMI) | Metallurgical change and confirmation that the installed alloy is the one specified | Supporting evidence; it does not measure wall or strain |
| MT or PT on welds and fittings | Surface cracking at welds, return bends and attachments | Surface methods; access and preparation set the coverage |
An example of the in-line category is Quest Integrity's Furnace Tube Inspection System, an ultrasonic tool that travels through serpentine and arbor coils and maps wall thickness and deformation along their length. Tools like this are run by their owners, and the decision to use one is usually about coverage: spot readings sample the coil, while a pig measures all of it.
\n\nThe radiant section is where creep and overheating live, so temperature history, diameter and wall carry the most weight there. The convection section is a different problem. Its tubes run cooler, are usually finned or studded, and sit behind the shock rows where access is poor. The usual threats are external corrosion under deposits and dew point corrosion in the coolest rows, so the plan focuses on the bare shock tubes, the accessible ends of each row and the tube sheets, with readings taken wherever cleaning makes them reliable.
\n\nRemaining-life and fitness-for-service assessments of heater tubes are engineering work that belongs to the owner's engineer. The inspection's job is to give that engineer temperature, wall and diameter data good enough to stand behind the result.
\n\nMost heater inspection problems are scoping problems: the wrong tubes measured, no baseline to compare against, or scale left in place so the readings cannot be trusted. Before the outage, settle these points with the inspection contractor:
\nAtlantis NDT mobilises crews for the shutdown and on-stream parts of the programme: infrared thermography, visual examination, ultrasonic thickness measurement and corrosion mapping, portable hardness testing, positive material identification, and MT or PT on welds and fittings. The work is planned and supervised by ASNT Level III personnel, and reports carry the procedure, calibration and certification behind each result. In-line coil pigging is run by its specialist vendors, and remaining-life assessment stays with your engineer. For the wider shutdown, see turnaround and shutdown inspection; for the exchangers on the same unit, see heat exchanger tube inspection and eddy current tube inspection. Quotes are prepared for each scope: send the heater data and outage dates.
\n\nThere is no single interval. API RP 573 guides the practice, but the frequency comes from the owner's inspection plan, set by the heater's service, its temperature history and the damage rates measured at earlier outages. Heaters that run hot or show coking trends warrant closer attention.
\nNo. Thermography tells you how hot the tubes ran and where, which is the key input for creep life. It does not measure wall thickness or swelling, so it directs the shutdown measurements rather than replacing them.
\nWhen the coil can be pigged and full coverage matters: localised process-side corrosion, a history of hot spots in unexpected places, or a remaining-life question that spot readings cannot answer with confidence. If the coil cannot be pigged, a denser spot measurement plan guided by thermography is the alternative.
\nBecause creep and short-term overheating depend on temperature and stress, not just thickness. A tube that ran well above its design metal temperature can swell and rupture even when its wall measures close to nominal, which is why diameter and temperature history sit beside thickness in the plan.
\nFurther reading: a field guide to API 571 damage mechanisms · fired heater (glossary) · all NDT inspection services.
","updatedAt":"2026-10-25"}