Ultrasonic testing authority for Freeport, Quintana Island and Port Freeport work

What a UT Level III actually approves, item by item

Ultrasonic testing is the method with the widest gap between what a certificate says and what the work demands. The Level III's approvals are specific and each one is a place the result can be invalidated: the written procedure and its essential variables, the calibration blocks and the reference reflectors in them, the sensitivity approach, the beam angles and probe frequencies, the scan plan and how coverage is demonstrated, the recording and evaluation levels, the acceptance standard invoked, and the practical examination that establishes an individual can execute it.

Essential variables deserve particular attention because they define when a procedure has to be requalified. Change the search unit frequency or element size, the wedge, the couplant, the surface condition, the material or thickness range, the scanning technique or the examination surface, and you may be outside what was demonstrated. Procedures that list essential variables and then never get revised as the equipment changes are the norm rather than the exception, and the equipment on a plant site changes constantly.

The Level III also owns the negative statement, which is the one most procedures omit. What will this procedure not find? A shear wave weld examination procedure written for planar defects has a known blind zone and a known sensitivity to reflector orientation. Stating that in the procedure protects everyone, because the alternative is an owner assuming a clean report means a clean weld.

Straight beam thickness and the arithmetic that goes wrong

Thickness measurement looks like the simplest thing in ultrasonics and produces more bad data than any other technique, because the errors are silent. Velocity is the first. An instrument set for carbon steel and used on austenitic stainless, aluminium, titanium or a nickel alloy returns a number that is confidently wrong. On a chlor-alkali or cryogenic site where alloys are mixed within a single unit, the procedure has to require velocity verification against a known step wedge of the correct material at the start of every shift and after any material change, with the check recorded.

Doubling is the second. A single-element probe has a dead zone, and below its minimum measurable thickness the instrument can lock onto the second backwall echo and display twice the actual wall. The reading looks plausible. It occurs precisely on the thinnest, most corroded point in the survey, which is the reading you most need to be right. Procedures should state a minimum thickness below which the technique is invalid and require a different probe or an A-scan verification.

Third is the coating. In single-echo mode the paint is included in the reading. On Gulf Coast lines that means a coating system of several mils reported as steel, on a line that may already be near retirement. The procedure must specify echo-to-echo or interface-to-backwall operation for coated surfaces and require the technician to verify the mode is active, not assume it. Temperature is the fourth: velocity falls as steel heats, so readings taken hot without correction read thicker than reality.

Angle beam sensitivity: DAC, DGS and transfer correction

Setting sensitivity is where the Level III's judgement is least substitutable. A distance amplitude correction curve is built by recording responses from identical reflectors at increasing sound paths in a reference block, then joining them. It is only valid for the block it was built on, the probe that built it, and material that behaves like the block. The reference block must match the part in material, nominal thickness, surface finish and heat treatment, and the reflector type and size must be what the code requires for the application.

Transfer correction is the step that gets skipped. The block and the part rarely have the same surface condition and attenuation, so a measured transfer loss must be established between them and added to the sensitivity. Skip it on a rough, as-welded or heavily attenuating component and every indication is being evaluated at less sensitivity than the procedure claims. Recording the transfer measurement in the report is what makes the claim verifiable months later.

DGS, sometimes called AVG, replaces the empirical curve with a calculated relationship between amplitude, distance and equivalent reflector size for a specific probe. It is fast and it is unforgiving of changes: swap the wedge, dress the probe face, or use a probe whose beam characteristics have drifted, and the curve no longer describes reality. A procedure permitting DGS must state how probe condition is verified and how often. Structural work under AWS D1.1 uses a different arithmetic entirely, rating indications as d equals a minus b minus c, where the attenuation factor c is derived from sound path. Technicians who move between ASME amplitude rules and D1.1 rating arithmetic in the same week make errors, and the practice should recognise that in how they are examined.

What UT certification does and does not cover

"UT Level II" on a card, with nothing else written, covers nothing that can be demonstrated. Certification under a written practice is method plus the techniques and scope that practice defines. A person examined on shear wave weld examination of carbon steel plate has not been examined on corrosion thickness surveys, on encoded corrosion mapping, on high temperature readings, on phased array sectorial scanning, or on time of flight diffraction. Those are different skills with different failure modes.

Limited certifications are a legitimate and underused tool. Certifying a technician for ultrasonic thickness measurement only, explicitly excluding flaw detection, is honest and defensible, and it lets a plant use people productively while being clear about the boundary. The alternative, which is what most sites do, is a general certificate and an informal understanding of who is actually any good at what. That understanding does not survive the auditor asking for the practical examination record.

Advanced techniques raise the bar further. Codes generally require procedure qualification by demonstration for encoded and advanced ultrasonic techniques, and the personnel performing them need documented technique-specific training beyond the base method. A defensible programme states this in the practice, examines against representative specimens, and keeps those specimens under control so the examination can be repeated and defended.

The practical examination, and the specimens nobody controls

The weakest record in most ultrasonic certification files is the practical examination. A line reading "practical: pass" with a date and an initial tells an auditor nothing. A defensible practical record identifies the specimens used, states the discontinuities present in them and their locations as known to the examiner, records what the candidate found and did not find, records the accuracy of length and depth calls where sizing is part of the scope, and applies a checklist with a stated pass threshold.

That requires controlled specimens, which is an asset management problem. Specimens must be identified, their known discontinuity maps kept confidential and retained, and their condition monitored, because a specimen that has been scanned by two hundred candidates over ten years accumulates surface damage that changes how it responds. Sites that borrow specimens from a vendor for exam day cannot produce this evidence afterwards.

Sizing deserves its own attention in the practical. Flaw sizing by tip diffraction, by amplitude drop, or from encoded data is a distinct competence, and the difference between a technician who can detect and one who can size accurately is exactly the difference between an indication that triggers a repair and one that is dispositioned by engineering analysis. If your programme uses ultrasonic sizing to support fitness for service work under API 579, the sizing competence must be examined and recorded, not assumed.

Where ultrasonics is the wrong tool, and saying so

A Level III who never says no is not providing authority. Several damage mechanisms common around Freeport are outside what conventional ultrasonics will find. External chloride stress corrosion cracking of austenitic stainless under insulation, driven by marine salt aerosol and wet insulation, is surface-breaking and finely branched; it is found after insulation removal by an appropriate surface method, not by a thickness grid. Reporting a clean thickness survey on that line gives false comfort.

High temperature hydrogen attack is the case with the worst history. It develops as methane bubbles and fissuring at grain boundaries in carbon and low alloy steels in hydrogen service above the relevant Nelson curve conditions, and it does not thin the wall until very late. A 2010 heat exchanger rupture at a Washington State refinery killed seven people and reset how the industry approaches detection. Screening for it requires advanced ultrasonic approaches, careful materials assessment against API 941, and a candid statement in the procedure of detection limits at early damage stages.

Material structure also imposes limits. Coarse-grained austenitic stainless and duplex welds scatter and attenuate the beam, and standard carbon steel techniques produce noise and missed defects. Low frequency dual-element transmit-receive probes and a procedure demonstrated on representative welded material are the minimum. Nine percent nickel plate in cryogenic service and aluminium cold box construction each carry their own velocity, attenuation and demonstration considerations. A procedure written for carbon steel and applied to any of these is a procedure that was never qualified for the job.

The Freeport ultrasonic workload

What UT actually gets asked to do in this county is unusually varied. Across the Brazosport chemical corridor, with Dow's Texas Operations complex and its Oyster Creek plant, Olin's chlor-alkali capacity and BASF's site, the daily load is thickness monitoring at established condition monitoring locations, corrosion under insulation follow-up, exchanger and vessel shell surveys, and weld examination on repairs and tie-ins. Alloy variety is the complicating factor: carbon steel, austenitic stainless, duplex, nickel alloys and titanium in wet chlorine service all appear inside a short walk of each other, and each has its own velocity and its own attenuation behaviour.

Freeport LNG on Quintana Island adds cryogenic construction and in-service work, where nine percent nickel plate and aluminium demand demonstrated procedures and where the consequence of a missed weld defect is not a leak into a bund but a brittle service failure. Inland, the Phillips 66 Sweeny complex represents the refining and hydroprocessing side of the county, where hydrogen service brings damage mechanism screening into scope.

Port Freeport and the barge and marine traffic on the Intracoastal Waterway bring a fourth category. Hull and tank thickness gauging for classification survey work runs on class rules, requires an approved service supplier as well as a qualified operator, and uses reporting formats and diminution criteria that have nothing to do with ASME. Shore-based NDT companies routinely underestimate this and lose the work at the qualification stage.

Engaging ultrasonic Level III authority

A typical engagement begins by reading the existing ultrasonic procedures against the work actually being performed, then walking two or three ultrasonic certification files backwards from the certificate to the practical examination record. That combination reveals almost everything: procedures that have drifted from the equipment in use, techniques being performed that no certification names, calibration approaches that would not survive a challenge, and practical records that cannot be defended.

From there the work is concrete. Rewrite or requalify the procedures that need it, with essential variables that match the equipment on site. Specify and verify the calibration blocks, including transfer correction requirements. Build examination banks and controlled specimen sets. Run the practical examinations and issue certifications within your written practice, naming techniques rather than issuing blanket cards. Where an owner wants a second opinion, provide independent review of ultrasonic data before it drives a repair or run decision. Nothing existing is stripped out; procedures and certifications that hold up are retained and properly referenced.

Anoop Rayavarapu holds ASNT NDT Level III certification in multiple methods, including ultrasonic testing, and is an API 653 Authorized Inspector. For a review of your ultrasonic procedures and certification files, or to discuss ultrasonic Level III authority for Freeport-area work, contact info@atlantisndt.com for a consultation.

Request a consultation