What a UT Level I Can Actually Do on a Beaumont Unit
A UT Level I in Beaumont works under a certified Level II or III. The level permits specific calibrations, specific tests and recording of results, and accept or reject calls only where a written instruction states the criteria. It does not permit choosing a technique or evaluating an indication independently. The employer's written practice defines the scope.
Beaumont sits at the top of the Sabine-Neches corridor, and the ultrasonic work there is dominated by thickness measurement on fixed equipment, weld examination on new fabrication and tie-ins, and corrosion mapping through turnaround windows. A Level I is the person who preps the surface, sets up the instrument against a reference block, takes readings on a defined grid and records them against a location scheme somebody else designed. That is not a small job. Most thickness data an owner will act on for the next five years is collected by Level I technicians, and a reading taken on scale, at the wrong temperature or against an uncorrected velocity is a corrosion rate that never existed. Training at this level is therefore weighted towards instrument control and repeatability rather than interpretation, because interpretation is not yours yet.
Source: ASNT SNT-TC-1A, sections 5, 6 and 8 and Table 6.3.1A; ANSI/ASNT CP-189; ASNT CP-105 topical outlines; ISO 9712:2021, Tables 1 and 2; ASME BPVC Section V, Article 1 (T-120), Article 4, Article 5 and Article 23; AWS D1.1 Clause 8; API 1104 section 11.
| Task on the job | UT Level I | UT Level II | What governs the split |
|---|---|---|---|
| Selecting the technique and writing the instruction | No | Yes, within an approved written procedure | SNT-TC-1A 5.2 and 5.3 responsibilities |
| Setting velocity and zero offset on a step wedge | Yes | Yes | Written instruction and procedure |
| Building a DAC curve on a side-drilled-hole block | Yes, exactly as the instruction states | Yes, and decides the primary reference level | Written procedure and code clause |
| Applying transfer correction between block and component | Applies the stated value | Determines whether the value is adequate | Procedure qualification |
| Deciding a corroded area needs a finer grid | No | Yes | Evaluation authority at Level II |
| Recording readings and instrument settings on the employer's form | Yes | Yes | Records requirement in the written practice |
| Interpreting an indication and evaluating it to a code | No | Yes | SNT-TC-1A 5.3 |
| Signing the report as technician of record | No | Yes | Written practice |
| On-the-job guidance of trainees | No | Yes | SNT-TC-1A 5.3 |
The sentence about Level I that almost everyone misreads
SNT-TC-1A describes a Level I as qualified to perform specific calibrations, specific nondestructive tests and specific evaluations for acceptance or rejection according to written instructions, and to record results. Read it slowly, because it does contain the words acceptance and rejection, and that is exactly why it is misquoted in both directions. One camp insists a Level I can never make a call. The other treats the clause as a licence to evaluate. Both are wrong. The clause grants the application of a criterion that somebody else wrote down for a condition somebody else anticipated.
The second half of the same clause is the half that gets skipped in the classroom: the Level I receives the necessary instruction and supervision from a certified Level II or Level III. Supervision is a defined relationship, not a feeling of being looked after. Your written practice has to say what it means on your site, who issues the instruction, whether the Level II must be in the work area, and who reviews and releases the results. When the practice is silent on this, the gap is not academic. It is the finding.
In day-to-day terms on a Beaumont unit, a Level I prepares surfaces, sets up and calibrates, takes readings on grids and welds that somebody else laid out, verifies calibration at the interval the procedure demands, and records everything so that another technician could repeat it. The scope is deliberately narrow. It is also where nearly all the raw data an owner will act on actually comes from.
Training hours, experience hours, and the difference between them
The recommended figures for ultrasonic testing at Level I are 40 hours of organised training, 210 hours of experience in the method and 400 hours of total nondestructive testing experience. Level II adds a further 40 hours of training and takes method experience to 630 hours and total experience to 1,200. Those numbers come from a recommended practice, and the employer's written practice can modify them. What it cannot do is modify them silently. Any departure has to be written down and approved by the certifying Level III, and an unwritten deviation is the easiest finding an auditor will ever record.
Training hours are not attendance hours. They have to follow a documented topical outline, delivered by someone qualified to deliver it, with the content, dates and hours recorded. Most employers adopt the ASNT topical outlines for this purpose. Experience hours are stricter still: they must be in the method, performed on production or representative work, and documented by task and date. Auditors compare logged training weeks against the job roster, and a block of forty training hours logged while the technician was signed onto a night shift two hundred miles away invalidates the whole file.
The ISO 9712 route counts time rather than hours. Ultrasonic testing at Level 1 calls for three months of industrial experience in the method, with further duration for Level 2, and a candidate going directly to Level 2 must accumulate the combined duration. A month is a nominal working month, not a calendar month with two weeks of leave in it, which is why documented experience and elapsed time so often disagree.
Velocity and zero offset: two unknowns, two known thicknesses
Every ultrasonic thickness measurement carries two unknowns. One is the instrument zero, which absorbs probe delay, wear face condition and cable. The other is the material velocity. A single-point calibration can only solve for one of them, which is why calibration on a step wedge or a multi-step block matters and why calibrating on a single known thickness leaves an error that grows or shrinks with wall thickness rather than staying constant. A Level I who understands this stops trusting a reading that looks plausible at the calibration thickness and drifts everywhere else.
Longitudinal velocity in carbon steel is about 5,900 metres per second, or 0.2320 inches per microsecond, and shear velocity about 3,230 metres per second, or 0.1270 inches per microsecond. The word about is where the trouble starts. Alloy, product form and processing all move the value. A block of mild steel used to calibrate for readings on stainless, on a clad vessel or on a nickel alloy transfer line introduces a systematic error that is invisible in the reading itself, because the instrument will happily display a wrong number to three decimal places.
For angle beam work the calibration block does more than set range. The IIW blocks establish the probe index point and verify the refracted angle, and the angle stamped on a wedge is nominal. Wedge wear, a repaired or re-glued wedge and the surface temperature of the component all move the real angle. Verifying it against the block, with the wedge you are actually using, is part of the setup, not an optional extra for exam day.
DAC and DGS: building the curve versus choosing it
A distance amplitude correction curve is built by plotting the response from identical reflectors, normally side-drilled holes, at increasing sound paths in a reference block matched to the component in material, thickness range and surface curvature. The curve compensates for beam spread and attenuation so that an indication at six inches of sound path can be compared with one at one inch. Transfer correction then accounts for the difference between the block surface and the real component, which on a corroded or rough Gulf Coast asset can be several decibels.
The DGS approach substitutes calculation for a reflector-bearing block. A probe-specific distance, gain and size diagram relates the echo from a disc-shaped reflector of a given equivalent diameter to sound path and gain. It is efficient and it removes a block from the truck, but it is valid only for the exact probe the diagram was produced for, in a defined condition. A worn wedge, a substituted probe of the same nominal frequency, or an unrecorded transfer correction invalidates the sizing without producing a single visible symptom on the screen.
The division of labour follows directly. A Level I builds the curve exactly as the written instruction states and re-verifies it at the interval the procedure demands. A Level II decides which reference level applies, whether the transfer correction is adequate for the surface actually encountered, and what the indication means against the code. That is not bureaucracy. The sizing rule and the acceptance rule come out of the same code clause, and only one of those levels is qualified to read it.
Couplant, surface condition and coatings
Sound will not cross an air gap, so the couplant is not a convenience. On a rough surface the operator's contact pressure changes the reading, roughness scatters energy and lifts the noise floor, and a thin film of the wrong couplant on a hot surface boils off between readings. Above roughly a hundred degrees Celsius the job needs a high-temperature couplant and usually a delay line, and on austenitic and nickel alloys the couplant has to meet halogen and sulphur limits because the residue left behind is a corrosion risk in its own right.
Coatings decide the technique. With single-echo measurement the paint is inside the reading, and a well-maintained asset in this corridor may carry several thousandths of coating that will be reported as steel. Echo-to-echo measurement times successive backwall echoes and reports steel alone, but it needs a coherent backwall. A rough, sloped or heavily pitted internal surface may never return a clean second echo, and a technician who keeps forcing gain to make one appear is manufacturing a reading rather than taking one.
Surface preparation has a trap of its own. Scale and loose rust must come off, but a technician who grinds a coupling flat into a corroded shell and then reports the thickness at the ground spot has reported the thickness of a location that no longer represents the vessel. Record what was removed, and where.
What the examination actually asks of you
Certification at Level I rests on three examinations. The general examination covers the principles of the method and is closed book. The specific examination covers the employer's equipment, procedures, specifications and the codes it works to, which is why it cannot be bought off the shelf and why it differs between two contractors working the same fence line. The practical examination is a demonstration on real specimens. The usual grading rule is a minimum of 80 percent on each examination and a composite of at least 80 percent, weighted as the written practice defines.
The practical is where candidates are surprised. The examiner hands over an instrument that is not set up, a written instruction, a set of specimens and a blank form, then works down a checkpoint list. Did you inspect the probe and cable. Did you calibrate in the sequence the instruction specifies rather than the sequence you prefer. Did you re-verify calibration at the stated interval and record it. Are the instrument settings on the sheet. Could another technician find your reading locations from your sketch. Most Level I failures are procedure and record failures, not physics failures.
Vision runs alongside all of it. Near vision acuity must be verified annually, commonly Jaeger Number 1 at not less than twelve inches or an equivalent, and colour contrast differentiation at the interval the written practice states. A missing annual vision record voids a certification that is otherwise faultless, and it is the single most common item pulled when an owner audits a contractor's personnel files.
Where the ultrasonic work is in Beaumont
The Beaumont and Port Arthur corridor along the Sabine-Neches Waterway carries one of the densest concentrations of refining and petrochemical capacity in North America, anchored by ExxonMobil's Beaumont refinery and complex and by the large refining and chemical plants immediately to the south. That base sets the ultrasonic workload: thickness monitoring on piping and pressure vessels, corrosion mapping in known damage locations, weld examination on turnaround tie-ins, and volumetric examination on new construction that has to be signed before it is insulated and lost from view.
Around that core sits a second market that a technician planning a career should not overlook. LNG export construction at Sabine Pass, methanol and ammonia production, tank farms and marine terminals along the waterway, the Port of Beaumont with its heavy break-bulk and military outload traffic, barge and vessel repair, and structural and pipe fabrication shops working to AWS D1.1 and ASME B31.3. Each of those brings a different code and a different acceptance rule for the same physical indication.
The practical consequence for a Level I is seasonal. Experience hours accumulate fast in the spring and autumn turnaround windows and thin out between them. Technicians who plan their documented hours around the turnaround calendar, and who make sure the hours are logged in the method rather than as general site support, reach Level II candidacy noticeably sooner than those who let the paperwork follow the work.
The findings that recur on Level I personnel files
Five findings account for most of what owners write up during contractor prequalification in this corridor. A missing or expired annual vision record. Training hours recorded without a topical outline, an instructor or dated attendance. Experience hours claimed in ultrasonics that the job roster shows were worked in another method or in no method at all. A technique sheet or written instruction that the supervising Level II never signed. And a certificate carried over from a previous employer and treated as current at the new one.
The last of those is worth stating plainly, because it costs crews their gate access every turnaround season. In the United States the certificate is the employer's statement that this individual meets that employer's written practice. It does not travel. The new employer may credit documented training and experience, but it has to satisfy itself, examine the individual under its own written practice and issue its own certificate before the technician performs work. Mobilising on the old card is not a lapse. It is work performed by someone the receiving employer never certified.
The recovery is unglamorous and quick if you start early: pull the vision record, pull the training outline and attendance, reconcile the experience log against the roster, get the instructions countersigned, and schedule the specific and practical examinations under the current employer's written practice before the mobilisation date rather than the week of it.
Getting from Level I to Level II
The step to Level II is not simply more hours. The additional 40 hours of training and the move to 630 method hours and 1,200 total hours are the entry ticket. What changes is authority: a Level II sets up and calibrates equipment, interprets and evaluates results against codes, standards and specifications, is expected to know the scope and limitations of the method, organises and reports results, and provides on-the-job guidance to trainees and Level I personnel.
That last responsibility is the one people underestimate. A newly certified Level II inherits the supervision obligation described in the written practice, which means the quality of the Level I data on that crew becomes their problem. The technicians who make the transition well are the ones who spent their Level I time understanding why the instruction said what it said, rather than only executing it.
A practical path in Beaumont: document your ultrasonic hours by method and date from the first week, ask for the specific examination scope early so you know which codes your employer actually works to, and get exposure to more than one geometry. A technician whose entire logbook is flat plate thickness on one unit is a weaker Level II candidate than one who has worked curved surfaces, welds and coated components, even at the same hour count.
Can a UT Level I sign off an acceptance decision?
Only inside a written instruction stating the criterion for the condition in front of you. SNT-TC-1A does allow a Level I to perform specific evaluations for acceptance or rejection according to written instructions, which is why the clause is often quoted as though it granted judgement. It does not. If the instruction says reject below a stated remaining wall and you find exactly that, you record the rejection. If you find something the instruction never anticipated, the decision is not yours and the report is not yours to sign.
How many training and experience hours does a UT Level I need?
SNT-TC-1A recommends 40 hours of organised classroom training, 210 hours of experience in ultrasonic testing and 400 hours of total nondestructive testing experience before initial certification at Level I. Those are recommendations, not law. Your employer's written practice is the controlling document and may modify them, but it must state the modification explicitly and the certifying Level III must approve it. Hours logged without a training outline, an instructor and dated experience records are not hours an auditor will accept.
What does supervision actually mean on a turnaround?
It means whatever your employer's written practice says it means, which is why a vague written practice is a liability. Some practices require a Level II physically present in the work area. Others accept a technique sheet issued and signed for the shift, with the Level II available and reviewing results before release. The distinction matters at two in the morning on scaffold when the geometry is not what the sheet drew. If the practice is silent, an owner's auditor reads the silence as an absence of supervision.
Why does a thickness reading change when the line is hot?
Sound velocity in steel falls as temperature rises, so an instrument calibrated on an ambient block over-reads on a hot component. The working correction is roughly a one percent increase in indicated thickness for every 100 degrees Fahrenheit above the calibration temperature, which you subtract. On a heavy wall that is trivial. On thin piping approaching minimum wall it decides whether the line stays in service, and it is the correction most often left out of a hand-recorded field sheet.
Does paint have to be removed before a thickness reading?
Not if you use an echo-to-echo technique, which times the interval between successive backwall echoes and therefore reports steel while ignoring the coating above it. Single-echo measurement includes the coating in the reading, and a well-maintained Gulf Coast asset can carry several thousandths of paint. Echo-to-echo has its own failure mode: a rough or sloped internal surface may never return a clean second echo, and knowing when to abandon it and prepare bare metal is a Level I skill.
What gets a Level I candidate failed in the practical examination?
Rarely physics. The examiner hands over an instrument that is not set up, a written instruction, specimens and a blank form, then records checkpoints: probe and cable verified, calibration performed in the sequence the instruction specifies, calibration re-verified at the stated interval, instrument settings recorded, readings tied to identifiable locations, sketch legible, form signed. Candidates who can find the flaw still fail for calibrating out of sequence or handing in a sheet nobody else could repeat.