Pipe Support Corrosion Inspection: Find the Gaps Around Trunnions and Supports

The blind spot is the geometry

The UK Health and Safety Executive's HID 1-2013 alert describes an LPG release from pipework carried on capped tubular trunnions. A weep hole sat directly against the supporting steel. Moisture entered the trunnion void, condensed on the process pipe and drove severe local wall loss where routine inspection could not see it. The pipe eventually ruptured under internal pressure. The incident had no connection to PetroBot.

HSE tells operators to include supports in the inspection regime and to use another means when an end cap blocks the view and moisture can enter. It also warns against permanently removing the cap without design work; the cap may provide stiffness. Treat that concealed interface as its own inspection zone from the start.

Credit evidence only to the area and method that produced it. An exterior photograph records visible configuration and condition. A qualified UT response records local thickness where the sound path was valid. Neither result describes pipe wall behind a cap, beneath a shoe or inside a contact face that the method could not reach.

Name the support before choosing the inspection method

‘Corrosion under pipe supports’ covers several geometries and several possible damage mechanisms. A tubular trunnion is a short tube welded to the process pipe and seated on steel. A shoe transfers load through an attachment and baseplate. Saddles spread contact; clamps and U-bolts restrain movement; guides control movement; resting supports create direct or lined contact. Caps, drains, welds, liners, orientation and load path all change the moisture path and the access available to an inspector.

Build the support register from current isometrics, support drawings and field photographs. Give every support a traceable ID. Record the line, service, pipe OD and material, orientation, support type, cap and drainage details, insulation or coating, operating temperature, environment, previous leaks or repairs and consequence. Then verify the arrangement in the field. A drawing will not show a blocked drain, missing liner, debris pocket or later modification.

The support name does not diagnose the damage. Coating failure, deposits, wet insulation, staining and poor drainage raise concern, but internal corrosion, erosion, external pitting, crevice corrosion and crack-like damage can overlap. Use the service history and field evidence to build a damage hypothesis, then choose methods that can test it.

Build a damage hypothesis before setting the scope

Set priorities from likelihood, consequence and detectability. For likelihood, check drainage design, cap and weep-hole position, wet insulation, coating age, temperature cycling, washdown or weather exposure, deposits, movement and the history of similar supports. For consequence, consider fluid, pressure, temperature, location, escalation potential and isolation difficulty. Detectability is the practical question: can the credible damage zone be seen, measured or not reached at all with the planned method?

Do not borrow a universal risk score, interval or scan pitch. Set those values from the owner's inspection program, the expected damage scale, the qualified method and the decision the data must support. Where the owner adopts API 570 and API RP 574, the responsible inspector applies the licensed requirements and local rules. PetroBot's acquisition plan must fit that program; it does not replace it.

Split the support into evidence zones

Draw the inspection boundary before mobilization. Separate the accessible adjacent pipe, support-edge transition, weld or profile, concealed contact or void and any coated, insulated or structurally obstructed area. Give each zone a technique, a data-validity rule and an owner for the next decision. ‘Near the support’ is only a location; it is not evidence that hidden wall was examined.

Expect the boundary to move in the field. Surface preparation can open a usable path. Curvature, roughness, a weld crown, unsafe access or loss of coupling can close one. Mark achieved coverage against the original zone map and record why it changed. Never interpolate adjacent thickness values across concealed wall to complete a map.

Use visual evidence to confirm configuration and condition

Run the visual survey systematically. Confirm the support type, cap and weep-hole arrangement, blocked drainage, coating breakdown, corrosion products, staining, debris, signs of wetting, deformation and the exposed surface around the attachment. Photographs should also show where insulation, steelwork, a weld profile or the support body blocks the planned route. This is the evidence you need to reconstruct the as-found condition and target follow-up.

A camera cannot quantify remaining wall behind the support. For every image, record the asset and support ID, viewing direction, location reference, date, visibility limits and any scale required by the procedure. Mark obscured areas as ‘not visible.’ Do not turn them into ‘visually acceptable’ areas in the report.

Accept UT only when the signal and location are valid

One-sided pulse-echo UT calculates local thickness from ultrasonic travel time. Trust the value only when the material and geometry are compatible, the surface and coupling are suitable, the probe and instrument setup are verified, the velocity basis is correct and the back-wall response meets the procedure. ISO 16809:2025 covers the principles of ultrasonic thickness determination. ISO 16831:2025 covers equipment checks for that work. ASTM E797/E797M-21 covers manual contact pulse-echo measurement and identifies limitations around non-parallel and concentric geometry.

HSE describes UT thickness measurement as a point measurement and ties measurement positions or grid pitch to the expected damage. Around a support, the procedure must separate a valid reading on accessible pipe from an unstable response near curvature or a weld, a rejected result caused by surface condition or coupling and a zone that was never interrogated. No single pitch will resolve every pit, support and service condition.

Coating, scale, roughness, curvature, temperature, pitting, probe seating and material response can all disturb the result. The approved procedure sets surface preparation, couplant, setup checks, repeatability and acceptance rules. If the response is inconsistent or the back-wall signal fails the rule, record it as rejected or no data and escalate. A displayed number is not automatically a valid thickness reading.

Use MagRover on the accessible ferromagnetic route

MagRover is PetroBot's magnetic crawler for ferrous surfaces, including curved, vertical and overhead pipe. It can collect camera evidence, UT spot readings, UT scans and location-correlated inspection data on a qualified route. PetroBot publishes 10 inches OD and above as the first pipeline screen. Final suitability still depends on the real surface and access conditions.

Before deployment, check pipe material and OD, curvature, route, coating and surface condition, temperature, magnetic adhesion, probe seating, access, couplant supply, power, tether management, recovery, inspection objective and site controls. Decide shutdown and surface preparation from those conditions and the governing procedure.

The working boundary is straightforward: MagRover collects evidence on suitable accessible pipe. A shoe, clamp, saddle, cap, insulation or cladding blocks that route unless the project has a specifically qualified access plan. Do not credit MagRover with wall inside a capped trunnion or behind support hardware. PetroBot supplies inspection evidence; the owner and qualified team set required thickness, calculate remaining life and decide acceptance, repair or fitness for service.

Escalate when adjacent robotic UT leaves the question open

Escalate when the important wall region remains concealed or the crawler route cannot be qualified. Do the same when images show trapped moisture or failed coating, readings shift abruptly, the back-wall response is unstable, rough geometry breaks coupling, crack-like damage is credible or the consequence makes the uncertainty unacceptable. A good scan on exposed pipe does not clear any of those triggers.

There is no automatic next technique. The owner, responsible inspector and NDT authority must choose another examination method, altered access or an engineered intervention that can answer the specific question. Base that choice on the suspected damage, material, geometry, one- or two-sided access, surface condition, sensitivity, method hazards, support loads and operating state. Keep screening, local characterization and engineering disposition as separate steps.

Record seven coverage states

Coverage must show what happened in the field, not make the final graphic look complete. Agree the reporting states, location convention and raw-data retention before mobilization. The seven states below form a practical owner-defined record; confirm which of them belong in the PetroBot contract and report template.

Keep the state separate from the evidence. ‘Attempted’ is not ‘measured,’ and ‘visual only’ earns no thickness credit. A rejected signal proves the method was attempted but failed its data-validity rule. A concealed zone remains an input to follow-up and residual-risk review until the responsible team closes it.

Run one controlled acquisition and data-quality workflow

Begin with an approved scope, a current support register and one shared location convention. Record the as-found condition, confirm the support configuration and mark the accessible boundaries. Verify the acquisition system under the governing procedure. Then link every image and UT result to the same asset, support, axial position and circumferential reference. Log route changes as they occur.

Assign a coverage state at each location and retain the reason. Repeat or expand an apparent low response only as the approved procedure directs. Keep the original and confirmatory records with surface preparation, coating, temperature, coupling, setup status and reviewer action. If the location, setup or historical conditions are not comparable, do not calculate a corrosion rate from unlike readings.

A decision-ready package identifies the line and support, drawing revision, scope objective, governing procedure and review roles. It carries the required equipment and setup records, planned and achieved coverage, valid UT with linked images, rejected data with reasons, concealed zones, field-condition notes and the handoff for complementary examination or engineering review. Put those requirements in the contract; do not assume every field is part of a standard deliverable.

Separate acquisition, interpretation and disposition

A valid UT result establishes wall thickness at one qualified location. A referenced image establishes visible condition from one view. Interpretation connects those facts with service, environment, geometry and history. Engineering disposition sets required thickness, corrosion rate, interval, expanded scope, repair, rerating, replacement or fitness for service. Keep the stages separate so the evidence and accountability remain visible.

Name the roles before the job. The owner, API 570 inspector where applicable, corrosion or materials specialist, NDT authority, piping or structural engineer and operations team each own different decisions. The integrity system should state who validates acquisition, interprets indications, approves access or support changes and accepts the final action. PetroBot's evidence feeds that process; it does not replace it.

Set the safety and recovery controls for the actual site

Build the task risk assessment around the process fluid and operating state, pressure and temperature, leaks and stored energy, area classification and work at height. Add the access equipment, dropped-object risk, exclusion zones, magnetic adhesion, electrical supply, tether route, couplant handling, communications, recovery and stop-work rules. Robotic access can move people away from some hazards, but competent operators, permits and manual verification still matter. Some supports will still need scaffolding or another access method.

Treat hazardous-area deployment as a configuration-specific approval. Verify the certificate and suitability of the exact robot, sensor, camera, tether, utilities and work method against the real area classification and site rules. Make the same project decision for live operation, shutdown, insulation removal and surface preparation; none follows automatically from choosing a crawler.

Stop for loss of adhesion, unsafe surface temperature, unstable coupling, inconsistent signals, an unplanned obstruction, a deteriorating process condition, a broken exclusion zone, uncontrolled couplant or loss of recovery. The inspection objective never overrides the site's safe system of work or the support's structural function.

Pipe-support corrosion inspection checklist

Define the decision and the blind spot before asking for a robot. List unknowns as survey risks instead of burying them as assumptions. Use this checklist for the suitability review, then let the responsible inspection team accept, revise or supplement the method before mobilization.

• What line, support register, isometric and support drawing define the asset and reconstructable support ID?
• What are the pipe OD, material, service, pressure and temperature context, orientation and operating history?
• Which support type, cap, drain, weep hole, liner, weld, contact face, coating and insulation details exist as found?
• Which damage hypotheses and consequence factors drive the inspection priority without being treated as diagnoses?
• Which zones are accessible adjacent pipe, edge transition, weld or profile, concealed contact or void, and obstructed area?
• What owner procedure, adopted code, personnel system, UT validity rules and acceptance authority govern the work?
• What surface preparation, adhesion, access, utilities, couplant, tether, recovery and area controls are available?
• How will planned, visual-only, valid, rejected, concealed, complementary and disposition states be recorded?
• Which visual media, UT data, setup records, coverage maps, exclusions and retained raw evidence are contractually required?
• What condition or coverage gap triggers another qualified method, changed access or engineering review?

Frequently asked questions

Technical references

• UK HSE HID 1-2013 — Corrosion in trunnion supports of hazardous-substance pipework
• UK HSE — Design Codes: Pipework
• UK HSE — Inspection and Non-Destructive Testing guidance
• API standards plan — API 570 and API RP 574 published editions and addenda
• ISO 16809:2025 — Non-destructive testing: ultrasonic thickness determination
• ISO 16831:2025 — Verification of ultrasonic equipment for thickness determination
• ASTM E797/E797M-21 — Manual ultrasonic pulse-echo contact thickness measurement
• ISO 9712:2021 — Qualification and certification of NDT personnel
• API Standard 570 — in-service piping inspection