Continuing Level III Authority for the Lower Mississippi Industrial Corridor

What of record actually means

A consultant who reviews a procedure and sends an invoice is not a Level III of record. The distinction is standing authority. The Level III of record is named inside the written practice as the individual who approves that practice, approves examination content, oversees the qualification and certification of personnel, approves the procedures those personnel work to, and is available to answer for the programme when someone external reads it. The name on the cover page is the whole point; an approval signature from someone with no defined standing in the document is not an approval.

Three roles get conflated constantly on Gulf Coast sites, and it is worth stating them separately. The API 510, 570 or 653 inspector authorises and signs the in-service inspection and sets the next interval — that is a certified appointment held by an individual and it carries the engineering call. The NDT technician is certified under the employer's written practice, to SNT-TC-1A or CP-189, to perform an examination exactly as a procedure directs. The ASNT Level III approves the procedures the technician works to, approves the practice under which the technician was certified, and interprets the code when a question arises. Three people, three sets of records, three different kinds of exposure.

What the role is not is equally important. A Level III of record is not a process safety management auditor, does not act as the API inspector of record, and does not substitute for the owner's own mechanical or corrosion engineering authority. Where those boundaries are blurred, the usual result is a document signed by someone whose certification does not support the signature, which is worse than no signature at all.

The industrial base this authority has to cover

The refining and chemical corridor along the lower Mississippi is the anchor. Refining runs at Norco in St. Charles Parish, at Garyville in St. John the Baptist Parish, and downriver at Chalmette and Meraux in St. Bernard Parish. Belle Chasse in Plaquemines Parish, once a refinery, now operates as a terminal — a reminder that in this region the asset base changes function faster than inspection programmes are rewritten. Large-scale chemical manufacturing sits alongside at Taft, Hahnville and Waggaman, with industrial gas plants and ammonia, methanol and chlor-alkali capacity spread along both banks.

The marine and fabrication economy is just as significant and follows different rules entirely. Shipyards and offshore support fabrication operate through the bayou parishes at Lockport, Larose and Houma; the Port of South Louisiana handles bulk tonnage at a scale few US ports match; the former Avondale yard now works as a logistics gateway; and liquid storage terminals cluster at St. Rose and along the river. Port Fourchon feeds deepwater Gulf operations. All of that brings ABS and Coast Guard expectations, thick-section and aluminium welding, and survey records that must reconstruct cleanly years later.

Then two outliers that most regional consultants are not equipped for. NASA's Michoud Assembly Facility in New Orleans East manufactures large launch vehicle structure, which brings aerospace personnel certification to NAS 410 and customer NDE specifications into a metro otherwise dominated by API codes. Upriver at Killona sits a nuclear generating station, which brings ASME Section XI inservice inspection and 10 CFR 50 Appendix B quality requirements. A Level III supporting this market has to know where each of those regimes starts and, more usefully, where it stops.

Who has jurisdiction, and what that means for your documents

Louisiana operates no state OSHA plan covering private sector employers, so federal OSHA has jurisdiction directly. For refineries and chemical plants that means 29 CFR 1910.119 process safety management, and specifically its mechanical integrity element, which requires inspection and testing on equipment in covered processes, performed by trained personnel, following recognised and generally accepted good engineering practice. In this industry that phrase resolves to API 510, 570, 653 and their supporting recommended practices. EPA's Risk Management Program under 40 CFR Part 68 sits over the top of much of the same equipment.

What is largely absent is a state inspector for unfired process vessels. The Office of the State Fire Marshal administers boiler safety under state law, but the pressure vessels and piping that carry the process are governed by the owner's own programme. The practical consequence is unusual and often underappreciated: the quality of your inspection is policed almost entirely by documents you author. There is no jurisdictional inspector arriving annually to catch a drifting programme, so the drift accumulates until an incident, an insurer, a client audit or a PSM compliance audit surfaces it all at once.

Other authorities overlay specific scope. PHMSA and the state pipeline authority govern hazardous liquid and gas lines under 49 CFR Parts 192 and 195, with their own personnel qualification expectations. The Coast Guard and the classification societies govern marine construction and repair, where approved welding and NDE procedures and survey-ready records are the deliverable. The NRC governs the nuclear station. Prime contractor NDE specifications govern aerospace structure. Several of these can bear on one operator at once, and the written practice has to be explicit about which applies where.

Turnaround season and the finding that always follows

The corridor runs on spring and fall turnaround cycles, and a large refinery turnaround can put several hundred contract NDT technicians on site inside a week. They arrive holding certificates issued under dozens of different employer written practices, of highly variable quality, with scope limits printed in small type that nobody reads at the gate. The work gets done, the unit starts up, and the records go into a box.

The finding lands later. A weld fails or a client audit samples the turnaround package, and three questions get asked that the host usually cannot answer. On what documented basis were the contractor certifications accepted? What were the scope limits on each card, and did the work performed stay inside them? Which procedure was each technician qualified against, and was it the procedure the report cites? Closing this out afterwards is expensive; a weld that cannot be shown to have been examined by a qualified person under an approved procedure is, for practical purposes, an unexamined weld.

The prevention is procedural and cheap by comparison. A written contractor certification acceptance procedure, a verification record completed before mobilisation rather than at the gate, a scope-limit check against the work packages actually assigned, and a short host-side practical verification for the techniques that carry the most consequence. That last item is unpopular and it is the one that finds problems. It is far better to lose two hours per technician at mobilisation than to re-shoot a hundred welds after startup.

Hurricane season produces a second, less predictable surge. Post-storm assessment, restart inspection and API 653 evaluation of storm-affected tankage all arrive at once, with the same contractor verification problem and less time to solve it. Sites that have the acceptance procedure already written handle it. Sites that do not, improvise, and the improvisation shows up in the records.

Gulf Coast damage mechanisms that shape the procedure library

Corrosion under insulation is the dominant wall-loss mechanism on this coast and the one most likely to be under-served by the procedure library. Heat, humidity, salt air and near-total insulation coverage combine with equipment that cycles through the water dew point, and the damage is invisible until the jacket comes off. A credible CUI programme needs more than one technique — profile radiography for screening, pulsed eddy current where insulation removal is impractical, targeted removal with UT prove-up, and a documented basis for which susceptible circuits get which. A single generic UT procedure does not constitute a CUI programme, however diligently it is applied.

The process-side mechanisms follow the crude diet and the unit. Sour and heavy crudes drive high-temperature sulfidic corrosion and naphthenic acid attack in crude and vacuum service, where alloy selection and sulphur and acid content interact in ways API RP 939-C describes and a thickness grid alone will not resolve. Hydroprocessing units bring high-temperature hydrogen attack, which is not a wall-loss problem at all and requires technique selection informed by the API RP 941 exposure history of each circuit. Amine and sour water systems bring wet hydrogen sulphide damage — blistering, HIC, SOHIC and sulphide stress cracking — with weld hardness and post-weld heat treatment history as the controlling variables.

Storage tankage deserves its own line because it is a substantial part of the corridor's asset base, sitting in terminals along the river and at the coastal facilities. API 653 evaluation is not just floor scanning. Magnetic flux leakage screening with ultrasonic prove-up of indications answers the floor question, but shell corrosion, settlement survey and the roof are part of the same assessment, and the settlement data is the part most often missing when the evaluation is reviewed. If the procedure library covers floor scanning and nothing else, the evaluation it feeds is incomplete before the engineer starts.

Nuclear and aerospace scope in the same metro

Commercial NDT programmes do not reach into either of these regimes, and pretending otherwise is a serious exposure. Personnel examining components under a nuclear station's inservice inspection programme qualify through performance demonstration under ASME Section XI rules, working inside a quality assurance programme that meets 10 CFR 50 Appendix B, with supplier requirements typically flowed down through ASME NQA-1. Employer certification under SNT-TC-1A is necessary background but nowhere near sufficient, and a written practice that implies otherwise creates a compliance problem the moment it is submitted.

Aerospace manufacturing runs on a different axis again. Personnel certify to NAS 410, procedures are approved against the customer's own NDE specification, and the surrounding quality system expectations — AS9100, and process approvals where they apply — are structurally unlike anything in a refinery. The paperwork discipline is higher and the tolerance for undocumented judgement is close to zero. A Level III moving between a refinery turnaround and a flight hardware programme has to change posture, not just paperwork.

For an organisation working across these markets — and in this metro plenty do — the written practice has to state the boundaries explicitly. Which methods and levels are certified under the site practice, which personnel additionally hold performance demonstration qualification and for what, which hold NAS 410 certification and under whose programme, and what work each is permitted to sign. Ambiguity here is not a documentation preference; it is how a technician ends up signing something they were never qualified to sign.

The arithmetic that quietly sets your next inspection interval

API 510 sets the internal or on-stream inspection interval for a pressure vessel at the lesser of one half the remaining life or ten years. That single rule means every error in your corrosion rate propagates directly into scheduling, and the direction of the error decides whether you over-inspect or run past the point you should have looked. This is why NDT data quality is an integrity question rather than a QA formality.

Two arithmetic traps recur. The first is computing a long-term corrosion rate from nominal thickness rather than the actual measured initial thickness, which quietly folds mill tolerance and any original overthickness into the rate. On a plate ordered with a positive tolerance the resulting rate can be materially wrong in the unsafe direction. The second is comparing a long-term rate against a short-term rate and taking the lower one because it looks more stable, when the short-term rate is higher precisely because something in the process changed. The code lets you use engineering judgement; it does not let you use the convenient number without recording why.

The third trap is not a corrosion rate at all. When a required thickness is recalculated for a vessel or a piping component, the joint efficiency used has to reflect the radiographic examination extent actually applied at construction. Assuming full radiography where only spot radiography was performed inflates the allowable and produces a required thickness that is too low. That original examination extent is a records question, which is exactly the kind of question an independent review of the inspection data set is for.

How the engagement is structured

The starting point is a document review rather than a proposal. Read the current written practice, the procedure library, a sample of certification files and a sample of completed reports, and produce a findings register ranked by audit exposure. That register tells both sides what the engagement actually needs to be, and in a fair number of cases it is smaller than the operator feared.

From there a Level III of record engagement is a standing arrangement written into the practice: the named individual, the methods their own certification covers, the scope of authority delegated to them, how procedure approvals are obtained between visits, the frequency of scheduled on-site presence, and a response commitment for the turnaround windows when a procedure revision cannot wait a week. Deliverables typically include written practice development and maintenance, procedure development and qualification, examination administration and certification oversight, technique selection for the damage mechanisms above, audit representation on technical questions, and independent review of inspection data.

Atlantis works a mobilised model rather than claiming a local office it does not have. Document work, procedure approval and remote review run continuously; on-site presence is scheduled around turnarounds, audits and examination cycles. Founder Anoop Rayavarapu holds ASNT NDT Level III certification in multiple methods and is an API 653 Authorized Inspector, which matters here mainly because it means the technical authority understands the inspection decisions your API inspectors have to make with the data. To discuss scope, contact info@atlantisndt.com and ask for a consultation.

Where the line is drawn

Atlantis supplies NDT technical authority. That means the written practice, procedure development and qualification, personnel certification within the practice, audit representation on technical questions, and independent review of inspection data. It does not mean acting as your API inspector of record — that appointment authorises and signs the in-service inspection, sets intervals, and belongs to a certified inspector inside or contracted to your organisation, working alongside the Level III rather than under them.

Atlantis is also not a process safety management auditor. Where a PSM or RMP compliance audit touches the mechanical integrity element, the Level III can supply the technical evidence, explain the procedure and certification basis, and answer questions on examination adequacy. Conducting the compliance audit itself is a different appointment with different independence requirements, and combining the two would compromise both.

Finally, a point worth stating plainly because it is asked often in this market: API 510, 570 and 653 inspector certification is administered by API, and preparing candidates for it is not part of this offer. NDT training — Level I, Level II and Level III preparation within the methods, and technique-specific training tied to your own practice and procedures — is a separate Atlantis service and is deliberately kept distinct from the technical authority role in your programme documents.

Request a consultation