UT method authority for Baytown refining, chemicals and steel
A UT Level III owns the ultrasonic method for your company — writing and approving the straight-beam, angle-beam, thickness and flaw-sizing procedures, setting reference sensitivity and DAC or DGS practice, and examining and certifying your UT technicians. Certification in UT is not certification in phased array, TOFD or any specific technique — each needs separate training and demonstration. Atlantis supplies that authority to Baytown operators and NDT contractors.
Baytown's ultrasonic workload is not generic weld inspection. It is thickness on ship channel terminal piping and API 653 tank shells, angle-beam on olefins and hydroprocessing welds, corrosion mapping on erosion-thinned elbows and deadlegs, wet H2S crack detection on sour service vessels, and advanced backscatter work where high temperature hydrogen attack is credible under API 941. Each of those is a different sensitivity, a different probe and a different acceptance basis, and a Level III who approves them all with one boilerplate procedure has approved nothing. The method certificate also has boundaries buyers misread. UT Level II certifies a technician to the employer's written practice for the techniques on which that technician was trained, examined and demonstrated. It does not confer phased array, time of flight diffraction, or the sizing skill an engineering critical assessment relies on, and it never substitutes for the API 510, 570 or 653 inspector who signs the inspection.
Source: Written against ASME BPVC Section V Article 4 (T-421 procedure requirements, T-431 couplant, T-434 calibration blocks, T-460 calibration) and Article 5 instrument linearity; ASME Section VIII Div. 1 provisions for UT in lieu of RT descended from Code Case 2235; API 510, 570, 653, API RP 571 and API RP 941; AMPP/NACE SP0296 for wet H2S cracking; ISO 16811 and ISO 17640 where DGS sensitivity setting is invoked; ASNT SNT-TC-1A and ANSI/ASNT CP-189.
| Decision | What the Level III approves | Typical failure when unapproved | Reference |
|---|---|---|---|
| Reference sensitivity | Basic calibration block, reflector type and depths, DAC construction, scanning gain above reference | DAC built on a block of the wrong thickness or curvature; scanning gain applied inconsistently between shifts | ASME V Art. 4, T-434 / T-464 |
| Transfer correction | Whether it is required, how it is measured, and how it is recorded on the report | Rough or coated examination surface loses several dB against a machined block; indications fall below reference and are never written down | ASME V Art. 4, T-464 |
| Instrument linearity regime | Screen height and amplitude control linearity checks, interval, and what invalidates data | Checks run once at purchase; a drifting instrument silently under-reports for months | ASME V Art. 5, T-463 and appendices |
| Technique release per technician | Which techniques a UT Level II certificate actually authorizes | Level II with weld experience put on high-temperature corrosion mapping with no demonstration | SNT-TC-1A Sections 8 and 9 |
| High-temperature scanning | Probe and wedge temperature rating, duty cycle, couplant, velocity and delay correction | Uncorrected readings on hot lines feed an optimistic corrosion rate into the next inspection interval | API 570 / ASME V Art. 4 |
| Sizing method | Whether amplitude drop, tip diffraction or a fixed technique is used, and its stated tolerance | Height reported to three decimals from a technique that has never been demonstrated to size anything | ASME V Art. 4, T-421 |
| Data retention | Instrument settings, calibration records, scan plans and raw data kept with the report | Report survives, evidence does not; the examination cannot be reconstructed during a dispute | ASME V Art. 1 and owner spec |
The ultrasonic workload Baytown actually generates
Ultrasonics on the Houston Ship Channel is not one job. The ExxonMobil Baytown complex spans refining, olefins and chemicals on a single site; Chevron Phillips Chemical's Cedar Bayou plant runs ethylene and derivatives; Covestro's Baytown Industrial Park makes polyurethane and polycarbonate precursors; JSW Steel USA runs a plate and pipe mill in the same city; and terminals, barge docks and fabrication yards fill the space between them. Each of those asks the ultrasonic method a different question.
On refining and chemical circuits the questions are damage-mechanism questions in the API RP 571 sense. Sulfidation and naphthenic acid corrosion produce general and localized loss where a thickness grid may or may not intersect the thin spot. Erosion-corrosion at elbows, tees and downstream of control valves produces a scalloped internal surface that defeats a spot reading. Wet H2S service produces hydrogen-induced cracking, blistering and stress-oriented HIC, which AMPP/NACE SP0296 addresses with a combination of straight-beam mapping and angle-beam crack detection — two techniques, two sensitivities, one asset. On older hydroprocessing and reforming equipment operating in the range where API RP 941 puts hydrogen partial pressure and temperature above the Nelson curve, high-temperature hydrogen attack becomes a credible mechanism, and detecting it needs advanced backscatter or full-matrix-capture techniques that a general-purpose weld procedure has no business claiming.
On the fabrication and terminal side the questions are construction questions: weld examination to ASME Section VIII or B31.3 acceptance, tank shell and floor work under API 653, structural welds to AWS D1.1. Different acceptance bases, different calibration, different reporting. One UT procedure covering all of it is a document nobody has read.
What Level III authority means for the ultrasonic method specifically
Every method has a decision that decides everything else. For ultrasonics it is sensitivity. Where reference level is set, how much scanning gain sits above it, whether transfer correction was measured on the actual part, and what fraction of a decibel of drift invalidates a shift's data — those four decisions determine what the examination is capable of finding, and they are all the Level III's to approve. Everything on the report is downstream of them.
The second decision is what the sound is being reflected off. ASME Section V Article 4 builds calibration on a basic calibration block with side-drilled holes at set fractions of the thickness, of the same product form and acoustically similar material and heat treatment as the part, with curvature constrained for small diameters and the block temperature held within a specified differential of the examination surface. Each of those constraints exists because violating it moves amplitude. A Level III who approves a procedure without naming the block, its material, its thickness band and its curvature band has approved a sensitivity that cannot be reproduced.
The third is what is done with an indication once it is found. Detection and sizing are not the same skill and not the same technique. Amplitude-drop sizing, tip-diffraction height sizing and encoded imaging give different numbers on the same flaw, and a fitness-for-service assessment under API 579 will be built on whichever number the report carried. The procedure has to say which method was used and what tolerance is claimed for it. Most do not.
DAC, DGS and the sensitivity decisions no one writes down
A distance amplitude correction curve is a physical claim about your specific instrument, probe, wedge, couplant and block. It compensates attenuation and beam spread so a reflector of a given size gives the same evaluated response wherever it sits through the wall. Under ASME Section V Article 4 it is constructed on the basic calibration block and confirmed at intervals — at the start and end of the examination, on any change of examination personnel, and at intervals not exceeding four hours. When a DAC point has dropped by more than the code's allowance in amplitude, or the sweep has moved beyond its allowance in distance, data acquired since the last valid calibration has to be re-examined or re-evaluated. That is not a suggestion; it is the mechanism that keeps a shift's readings meaningful.
DGS works from a different premise: the probe manufacturer's distance-gain-size diagram, a reference reflector such as a backwall or side-drilled hole, plus entered values for material attenuation and transfer loss. It is efficient, it needs less block hardware, and it is standard practice where ISO 16811 and ISO 17640 govern. It is also not what ASME Section V Article 4 describes. Contractors who learned DGS on European work and carry it onto an ASME purchase order create a finding that is difficult to argue with, because the code text is explicit about how sensitivity is established.
Transfer correction is the piece that quietly disappears from both approaches. The calibration block is machined; the pipe is corroded, painted, or ground. The difference can be several decibels, which is the difference between an indication that gets recorded and one that never appears. If the procedure does not state how transfer correction is measured and where it is recorded, assume it is not being done.
What a UT certificate does not cover
A UT Level II certificate says the holder was trained, examined and demonstrated competent, under a named employer's written practice, in the ultrasonic method. It does not say the holder can run a phased array weld inspection, interpret a TOFD image, encode a corrosion map, apply high-temperature techniques, or size a crack for a fitness-for-service assessment. Those are techniques within the method, and a defensible written practice states which techniques each certificate covers.
This distinction is where owners and contractors most often disagree in Baytown, because the commercial pressure runs one way. A turnaround needs bodies, a Level II with fifteen years of weld experience is available, and the scope is corrosion mapping on hot, insulated piping. The certificate is valid; the assignment is not covered by it. The Level III's job is to say so before the data exists, and to run the training and practical demonstration that makes the assignment legitimate — not to bless it retrospectively.
There is a second boundary. UT certification is an NDT qualification, not an inspection authorization. The API 510, 570 or 653 inspector authorizes and signs the in-service inspection and owns the continued-service decision. The NDT technician acquires data under the employer's written practice. The ASNT Level III approves the procedure and the practice. Three distinct roles, and Atlantis occupies exactly one of them: it supplies NDT technical authority, not inspector-of-record services and not PSM auditing.
Errors that reach the corrosion rate
Thickness data is not read as thickness; it is read as a trend, and the trend feeds an inspection interval. That amplifies small measurement errors into scheduling errors. Temperature is the classic case: sound velocity in steel falls as temperature rises, so a gauge configured with an ambient velocity reads thick on a hot line by roughly one percent per hundred degrees Fahrenheit above ambient. Take a 0.500 in. nominal wall at 400°F: uncorrected, it can read about 0.015 in. thick. Compare that against a previous reading taken cold during a shutdown and the circuit appears to have gained metal, or, worse, appears to be corroding slowly when it is not.
The second amplifier is resolution against interval. A gauge resolving to 0.001 in., used on a wall losing 0.004 in. per year, read at a 24-month interval, produces a corrosion rate whose measurement uncertainty is comparable to the rate itself — before you add repositioning error at the CML, couplant variation, and whether the same technician found the same spot. API 570 distinguishes short-term from long-term corrosion rate precisely because a single interval is not trustworthy, and a procedure that does not specify CML marking, repeat readings and minimum-of-N recording is handing the integrity engineer noise.
The third is the backwall itself. Erosion and erosion-corrosion produce a scalloped, sloped internal surface. A single-element probe on a sloped backwall either loses signal entirely or returns the strongest reflector in the beam, which is not the thinnest point. Dual-element probes, lower frequency, and encoded B-scan or corrosion mapping exist to solve this, and choosing between them for a given circuit is a Level III decision, not a technician preference.
UT in place of radiography on a live unit
A great deal of Baytown ultrasonic work exists because radiography does not fit the site. Shooting a source on a live ship channel unit means clearing a boundary, coordinating with operations, working nights, and operating under a Texas Department of State Health Services radioactive materials license with all the survey, logging and RSO obligations 25 TAC Chapter 289 attaches to it. Ultrasonics removes the exclusion zone and lets the work proceed alongside the plant.
That substitution is legitimate and code-supported — modern ASME Section VIII Division 1 rules descended from Code Case 2235 permit ultrasonic examination in lieu of radiography for certain welds — but the conditions are exacting. The technique must be demonstrated, the personnel qualified beyond a basic Level II certificate, the scan plan capable of covering the volume, and the acceptance criteria flaw-based rather than amplitude-based, which changes what the report must contain. A contractor who says "we'll UT it instead" without meeting those conditions has not substituted an examination; they have skipped one.
This is the single most common place a second opinion pays for itself. Before a fabricator commits a schedule to UT in lieu of RT, someone with method authority should confirm the geometry is scannable, the demonstration block exists or can be made, and the personnel qualification route is real. Discovering otherwise after the welds are complete is expensive in a way that cannot be recovered.
Certifying and re-releasing UT technicians
Ultrasonic examinations produce more findings on personnel files than any other method, because ultrasonics has the widest gap between what a certificate implies and what a technician can actually do. The specific examination must be genuinely specific — questions on the procedures, equipment and materials your company works with, not a generic question bank that would be identical at any employer. The practical must place the candidate in front of representative hardware with representative flaws and be graded against a written checklist that survives inspection.
Records are the other half. Training hours in the method with dates and content, experience hours substantiated rather than asserted, near-vision acuity on a Jaeger #2 or equivalent chart at not less than twelve inches at the interval your practice specifies, and a recertification interval that the practice defines and the company actually observes. Where CP-189 has been invoked alongside SNT-TC-1A, the stricter requirements govern — including that the Level III hold ASNT Level III certification — and most companies invoking both have not noticed.
Atlantis prepares and grades those examinations, verifies the underlying records, documents technique-level releases so the certificate's real scope is unambiguous, and stays named in the practice so the next audit has someone to talk to. To scope UT method authority for your operation, contact info@atlantisndt.com for a consultation; engagements are quoted on request.
What exactly does a UT Level III approve?
The written procedure and its revisions; the calibration approach, including the basic calibration block, reference reflectors and DAC construction; scanning sensitivity, transfer correction and scan overlap; the sizing method and how indications are recorded; the instrument linearity check regime; and the training, examination and practical demonstration behind every UT technician's certificate. He also decides which techniques a given technician is released to perform, which is narrower than the certificate implies.
Does UT Level II certification cover phased array and TOFD?
Not automatically, and treating it as though it does is one of the most common findings on the Ship Channel. UT is the method; phased array, TOFD, corrosion mapping, high-temperature thickness and creeping-wave techniques are applications within it. Your written practice has to state which techniques each certificate covers, and the Level III has to have trained, examined and watched a practical demonstration on that technique before releasing the technician to it.
Can we use DGS curves instead of building a DAC?
Only where the referencing code allows it. DGS is an ISO-lineage practice — sensitivity set from a probe's published diagram plus a reference reflector, attenuation and transfer loss. ASME Section V Article 4 is built around a basic calibration block with side-drilled holes and a distance amplitude correction curve. If your customer's purchase order invokes ASME, a DGS setup is not a drop-in substitute, and using one without written agreement will be written up.
Why do our thickness readings disagree between two technicians?
Usually velocity, temperature and backwall condition rather than skill. Steel velocity falls with temperature, so an uncorrected reading on a hot line reads roughly one percent thick per hundred degrees Fahrenheit above ambient. A single-element probe on an eroded, scalloped backwall reports an average, not a minimum. Different couplant and surface prep move amplitude. The procedure, not the technician, is supposed to close those variables, and that is the Level III's job.
Is API 510, 570 or 653 inspector training part of this offer?
No. Atlantis does not run API inspector certification programs and a UT Level III engagement produces no API credential. The two roles sit side by side: the API inspector signs the in-service inspection and owns the fitness-for-service judgment; the UT Level III owns whether the ultrasonic data underneath it was acquired to an adequate procedure by a properly certified technician. Buyers who blur them end up with neither authority properly covered.
Which Baytown assets most often need a second UT opinion?
Older hydroprocessing and reforming circuits where high-temperature hydrogen attack is credible under API 941; sour service vessels and piping with wet H2S cracking history; erosion-thinned elbows, tees and injection points on olefins and utility lines; tank shells and critical zones on API 653 inspections; and any weld where UT was substituted for radiography because the unit could not be cleared for a shoot.