Internal Pressure Vessel Inspection in Malaysia | RVI and UT

The technical thesis: correlate what the camera sees with what UT measures

Remote visual inspection and ultrasonic thickness measurement answer different questions. RVI can document deposits, corrosion products, coating or lining condition, erosion patterns, deformation, weld condition and the physical context around a location. Contact UT estimates remaining wall from acoustic time of flight at the probe position. Neither dataset is complete without traceable location information, and neither should be stretched beyond its qualified purpose.

API 510 addresses the in-service inspection, rating, repair and alteration of pressure vessels, while API RP 572 provides inspection-practice guidance including planning, methods, limitations, records and follow-up. These documents do not turn a robot into an authorized inspector. They establish why the robotic work package must sit inside the owner's inspection program, with the responsible inspector defining scope and an engineer evaluating thickness against the applicable construction code and service conditions.

A defensible robotic workflow therefore creates one coordinate framework for the vessel, records RVI before cleaning away useful evidence where practicable, establishes a planned UT grid, confirms anomalous readings and preserves inaccessible areas as explicit exceptions. The output is an auditable condition dataset for the inspection team, not an automatic return-to-service decision.

Start with the vessel, service and credible damage mechanisms

The grid should be designed from how the vessel could deteriorate, not from a convenient square spacing alone. Service chemistry, temperature, pressure cycles, phase level, flow direction, internals, metallurgy, fabrication details, lining or coating, previous repairs and historic inspection findings all change where damage may concentrate.

Broad internal corrosion may justify a representative grid over shell courses and heads. Inlet impingement, wet-gas condensation, liquid-vapor interfaces, sediment, water bottoms, crevices and drainage low points can produce more localized loss that needs tighter targeted coverage. Weld seams, nozzles, manways, attachment zones and repaired areas may require separate techniques or manual prove-up because robot motion and probe seating are more difficult there.

UT thickness does not constitute a general crack examination. Stress-corrosion cracking, fatigue, hydrogen damage, blistering, laminations and weld flaws require damage-mechanism-specific methods and qualified procedures. RVI may show a surface-breaking symptom, but absence of a visible indication is not evidence that a crack-like mechanism is absent.

• Design basis: vessel drawings, materials, nominal thickness, corrosion allowance, design conditions and weld map
• Operating history: contents, contaminants, phase levels, temperature, pressure cycles, upsets and cleaning history
• Damage history: prior TMLs, corrosion rates, repairs, alterations, leaks and previous inspection limitations
• Geometry: shell courses, heads, nozzles, manways, seams, internals, supports, attachments and low points
• External context: burial, coating, cathodic protection, backfill, groundwater and any previous excavation results

RVI and UT gridding provide different evidence

ASNT describes remote visual testing as viewing an otherwise inaccessible area using equipment such as a camera, borescope or video probe. Inside a pressure vessel, the value of RVI is not merely live video. A useful record identifies the viewed surface, orientation, lighting, scale, camera distance where relevant, visible condition and any zones the camera could not adequately resolve.

ASTM E797/E797M covers manual contact pulse-echo thickness measurement where only one side of a component is accessible. This principle is directly relevant inside a vessel: the probe can measure remaining wall from the internal surface when material, surface, coupling and calibration conditions are suitable. A grid organizes those discrete readings spatially; it does not turn them into measurements of every point between grid intersections.

Where MagRover RVI and UT fit

PetroBot's published service and product material positions MagRover as a magnetic crawler for accessible ferromagnetic pressure-vessel surfaces, with a camera mount for visual inspection and UT spot-reading or scanning workflows. It also identifies power and water for couplant as deployment requirements. For internal work, this means a prepared, out-of-service vessel with an entry opening and a continuous route on which magnetic adhesion, camera visibility and acoustic coupling can be demonstrated.

The operator can control the robot and review the feed from outside the manway while the crawler traverses suitable shell or head areas. That can reduce the amount of time personnel spend inside and can make grid acquisition more traceable. It does not prove that the entire vessel can be inspected without entry: staging, cleaning, obstruction removal, manual confirmation, retrieval or examination of complex details may still require qualified personnel under the site's entry program.

Magnetic attraction is only the first go/no-go test. Scale, sludge, product residue, non-ferromagnetic linings, thick coatings, weld caps, reinforcing pads, sharp transitions, internals, small radii and discontinuities can impede travel or prevent stable probe contact. A representative trial must verify the whole system - traction, tether behavior, video, lighting, position control, couplant delivery and repeatable UT response - before planned coverage is credited.

Underground pressure vessels: valuable internal data with a strict boundary

A buried or mounded pressure vessel presents a practical access problem because most of the external shell is unavailable without excavation. When the vessel is removed from service, isolated, cleaned and safely accessible through a manway, internal MagRover RVI and UT gridding can provide position-linked condition and remaining-wall data on reachable ferromagnetic areas. Measuring from the internal side is physically valid for one-sided pulse-echo thickness work when the procedure and surface conditions are suitable.

The interpretation boundary is crucial: a thickness reading reports the acoustic distance between the internal and external surfaces at that location. It does not identify which surface caused the loss. Internal video may support an internal-corrosion interpretation where corresponding morphology is visible, but an apparently sound internal surface does not rule out soil-side corrosion, coating failure or cathodic-protection problems.

For an underground vessel, the internal dataset should therefore be reviewed alongside external-corrosion controls and evidence: coating specification and age, cathodic-protection criteria and survey history, groundwater and drainage conditions, backfill, leak history, previous excavations and any above-grade appurtenance findings. Targeted excavation or another qualified method may still be required when external degradation is credible or a low reading cannot be attributed.

Deployment prerequisites and go/no-go checks

Internal robotic inspection begins only after the owner has made the vessel safe for the planned work. The scope should state whether any person will cross the entry plane, because remote robot operation from outside does not cancel permit-space obligations for personnel who enter. Isolation, depressurization, draining, cleaning, ventilation, atmosphere testing and rescue arrangements remain site responsibilities under the applicable safe-work system.

The robot, camera, lighting, UT instrument, probe, couplant delivery, position reference, tether and retrieval arrangement form one inspection system. Qualifying the thickness gauge alone is not enough. A pre-job review should compare the vessel drawing with crawler dimensions and turning needs, then test a representative surface or mock-up where uncertainty exists.

• Approved inspection plan, grid drawing, acceptance route and named responsible inspector or engineer
• Vessel out of service, depressurized, drained, positively isolated, cleaned and atmosphere-controlled
• Confirmed ferromagnetic base material and accessible route across the intended shell and head zones
• Manway size, internal obstructions, weld profiles, surface curvature, lining and surface condition reviewed
• Camera lighting, focus, orientation reference and image-storage check completed
• Qualified UT procedure, reference standard, material velocity, thickness range and probe configuration
• Power, water or approved couplant supply, containment and equipment-suitability review completed
• Tether management, communications, loss-of-adhesion response and recoverable robot retrieval plan

Design the grid around geometry and expected damage

A vessel grid needs a reproducible datum. For a horizontal vessel, one practical convention uses axial distance from a named tangent line or weld and circumferential clock position viewed from a specified end. For a vertical vessel, elevation and circumferential angle can reference a fixed nozzle or north mark. Heads need a defined polar or meridional convention rather than forcing shell coordinates onto a curved surface.

There is no universal grid spacing. The responsible inspector selects it from the credible damage morphology, previous results, required confidence, probe footprint, geometry and applicable inspection plan. Broad general corrosion can be represented differently from narrow inlet erosion or pitting. When a low or abrupt gradient is found, the procedure should require tighter local gridding or a qualified scan until the boundary and minimum are adequately characterized.

Acquire RVI first, then UT without losing traceability

Where safe and useful, begin with an as-found RVI pass before final surface preparation. Record deposits, corrosion products, staining, lining damage and debris that cleaning could remove. Use consistent lighting and deliberate camera motion, capture overlapping views, pause for still images, and include a known feature or scale when size matters. Every clip or image should link to the vessel datum, zone and travel direction; an unlocated video is difficult to use in a future comparison.

After cleaning to the procedure's required condition, verify the UT system before production readings. Contact UT converts acoustic transit time using an assumed material velocity, so surface roughness, poor probe seating, curvature, insufficient couplant, coating or lining echoes, material condition and incorrect calibration can bias the result. ASTM E797/E797M emphasizes a qualified technique and experienced interpretation rather than treating the displayed number as self-validating.

During acquisition, log planned, measured, rejected and inaccessible cells separately. Confirm unexpectedly low or unstable readings from another direction or with an approved manual technique. Perform post-run verification before changing the setup. A color map is a visualization of accepted data, not the primary evidence, and interpolation must not conceal unmeasured gaps.

• Pre-run and post-run calibration or verification checks within the written procedure's tolerance
• Reference standard, probe, cable, couplant, velocity, zero and instrument settings recorded
• Grid marks or encoded distance checked against a physical vessel reference
• Signal-quality review for every low, abrupt change, dropout or geometry-affected reading
• RVI media and UT values linked to the same zone, coordinates and orientation
• Raw data, accepted readings, rejected readings, processing settings and exception log retained
• No-data and inaccessible areas displayed distinctly from measured or interpolated areas

Interpret the combined dataset without overclaiming

Interpretation begins by correlating evidence. A visual corrosion zone with a corresponding UT gradient is stronger evidence than either record alone, but the report should still separate observation from inference. For buried vessels in particular, reduced thickness with no visible internal morphology may be consistent with external loss, a material or coupling issue, or a cleaned internal surface; it is not proof of any one cause.

A minimum reading needs context: exact coordinate, raw signal quality, surrounding values, visual condition, nominal or baseline thickness and confirmation status. Do not calculate a corrosion rate unless current and previous readings are truly comparable in position, surface, technique and calibration basis. Where old spot locations cannot be reconstructed, the robotic grid may be best treated as a new baseline.

The authorized inspector and responsible engineer determine acceptability under the governing inspection and construction codes. Minimum required thickness, corrosion allowance, remaining life, rerating, repair and fitness-for-service require design conditions, materials, loads and flaw geometry beyond the crawler's output. Crack-like features, blistering, suspected hydrogen damage, severe local attack or inconsistent responses should trigger the appropriate complementary NDE and engineering review.

Safety and operational controls remain vessel specific

A pressure vessel may meet the definition of a confined space, and OSHA specifically lists vessels as examples. Whether it is a permit-required space depends on the hazards and local regulatory framework. Keeping the operator outside the manway can reduce exposure, but if anyone enters for cleaning, staging, confirmation or recovery, the site's entry permit, isolation, atmospheric monitoring, attendant, communication and rescue provisions still apply.

The complete deployed equipment configuration must be reviewed for the location and residual hazards. Electrical power, lighting, water or couplant, tethers and cleaning equipment can introduce ignition, slip, trip, contamination and retrieval risks. Equipment suitability must be established from applicable documentation and site classification; this article does not claim a hazardous-area certification for a particular MagRover configuration.

Define stop-work triggers before the run: loss of atmosphere control, unexpected residue, robot instability, tether damage, loss of communications, excessive leakage or couplant accumulation, evidence of severe wall degradation, or any condition outside the approved inspection procedure. A recovery plan should retrieve the crawler without creating an improvised human entry.

A decision framework for the next action

Combine data confidence, coverage and condition severity. This prevents a clean-looking but incomplete visual record from closing a risk, and prevents one low-confidence UT value from automatically driving repair. Acceptance thresholds and response times should be defined by the owner and responsible inspector before acquisition where practicable.

Practical checklist for an auditable vessel inspection

A useful deliverable lets another competent reviewer reconstruct the scope, location, evidence and limitations. If one answer below is missing, record it as an uncertainty and decide whether the remaining data can still support the intended integrity decision.

• Are vessel identity, service, materials, drawings, nominal thickness and inspection history confirmed?
• Which credible damage mechanisms and integrity decisions determine RVI scope and grid density?
• Is the vessel isolated, cleaned and released under the site's safe-work and entry controls?
• Is MagRover suitability verified for ferromagnetism, access, curvature, lining, obstacles and retrieval?
• Are shell and head coordinates tied to drawings and an unambiguous physical datum?
• Do RVI media preserve orientation, lighting, scale where needed and as-found surface context?
• Do UT calibration, position, signal-quality and pre/post verification records bracket the readings?
• Are low readings confirmed and measured, rejected, inaccessible and interpolated areas distinguished?
• For an underground vessel, are internal results correlated with coating, cathodic-protection and soil-side evidence?
• Does the report separate observations, measurements, technical limitations and engineering disposition?
• Are follow-up NDE, repair, monitoring or excavation actions assigned to an owner and due date?

Frequently asked questions

Technical references

• API - API 510 Pressure Vessel Inspection Code, 11th Edition announcement
• API - API RP 572 Inspection Practices for Pressure Vessels, 5th Edition
• API - API RP 571 Damage Mechanisms Affecting Fixed Equipment
• ASTM E797/E797M-21 - Manual Ultrasonic Pulse-Echo Contact Method
• ASNT - Visual Testing, including remote visual testing
• OSHA 29 CFR 1910.146 - Permit-Required Confined Spaces
• HSE - Pressure systems: written schemes of examination