Study the IWI subject matter by practicing a four-step decision chain on paper cases: code the imperfection, quote the specified quality level, apply the stated limit to the measurement, and write the disposition. Drill each step separately before combining them, and verify your reasoning with a scoring rubric rather than your memory of limits.
What the B, S and C tiers actually separate in one IWI framework
The International Welding Inspector credential is one qualification family with three tiers of increasing technical breadth and responsibility. The core subjects are shared; the depth of judgement expected from you scales with the tier you hold.
IIW maintains a personnel qualification and certification framework that includes the International Welding Inspector alongside the engineer and specialist routes. The inspector credential is offered at basic, standard and comprehensive levels, which means your preparation should start from the shared core: welding processes, joints, imperfections, inspection methods, and quality documentation. Do not treat the tiers as three different subjects to memorize three times.
The practical study consequence is scope-setting. When you practice a case, decide in advance what your tier is expected to do: recognize an issue and refer it upward, or decide and justify the decision yourself. Write every practice answer in the form identify, then refer or decide, then justify. This habit keeps your reasoning depth matched to your tier; rehearsing recognition alone leaves the justification skill undeveloped at tiers that require it, and rehearsing only open-ended justification at a tier that expects referral produces answers that overstep the role. Matching depth to tier is what makes tier-specific study efficient rather than redundant.
- One credential family, three tiers: basic, standard, comprehensive.
- Shared core subjects; scale of expected judgement differs by tier.
- For every practice case, state explicitly: identify, refer or decide, justify.
Naming and coding imperfections before judging them
Describing and judging an imperfection are two separate tasks. ISO 6520 gives each imperfection type a number and definition so inspection reports are unambiguous; acceptance judgement belongs to a different standard with quality levels.
ISO 6520 is a classification standard: it assigns reference numbers to imperfection types so a report written in one country reads identically in another. Learn the structure by series: cracks in the 100 series, cavities such as gas pores in the 200 series, solid inclusions in the 300 series, lack of fusion and penetration in the 400 series, and shape and dimension defects such as undercut in the 500 series. Examples worth drilling early include undercut as 5011 and gas pore as 2011, plus arc strike and spatter from the 600 series because they appear on real components even when the weld itself is sound.
Train a strict habit on every case: before any accept or reject thought, write the code, the name, the location, and the size. A report entry such as 5011 undercut, 60 mm from tack, 0.8 mm deep, 60 mm long is a complete observation. This discipline matters because identification and disposition are graded as separate skills: if you fuse the two steps, you tend to jump straight to a verdict and the descriptive half of the answer goes unstated, even when the verdict itself is correct.
Which standard answers which inspection question
Preparation becomes manageable once you map each recurring inspection question to the standard family that answers it. Build this mapping yourself rather than memorizing isolated facts, because case questions always arrive embedded in one of these decision types.
The table below organizes the core decision types an inspector faces. Notice that acceptance judgement for steels and for aluminium are separate standards, and that personnel and procedure approval are also separate: a welder certificate and an approved welding procedure are two independent documents. Keep this separation in mind whenever a case presents paperwork, because mixing the two routes is a genuine conceptual trap rather than a careless slip.
Use the table as a self-test: cover the right-hand columns, read the question, and name the standard family aloud. Any hesitation marks a topic for targeted review. Then go one level deeper for each row by asking what the standard actually delivers, for example a classification code, a limit table tied to quality levels, or a qualification range you must check against production parameters.
| Inspection question | Standard family | What it delivers | Typical inspector use |
|---|---|---|---|
| How do I name and number this imperfection? | ISO 6520 | Classification and numeric codes for imperfection types | Unambiguous report wording |
| Is this steel weld acceptable, and at what quality level? | ISO 5817 (steels); ISO 10042 (aluminium) | Quality levels D, C, B with limit tables | Accept or reject against the specified level |
| How do I examine visually and record results? | ISO 17637 | Methods, aids, and documentation practice for visual examination | Structured before, during, and after welding checks |
| Is this welder approved for this work? | ISO 9606 series | Welder qualification ranges (material, joint, position) | Personnel verification against the job |
| Is this WPS supported by approved qualification data? | ISO 15614 series | Procedure qualification ranges for welding variables | Trace WPS to WPQR and check variable ranges |
Scenario 1: judging undercut with the right quality level
A described butt weld with undercut tests whether you apply the limit from the specified quality level and thickness, not from intuition. The decision itself is unforgiving in a specific way: a measurement compared against a remembered limit for the wrong level yields a confident but wrong verdict.
Paper case: a butt weld in 12 mm carbon steel plate. The contract specification requires quality level C. Your case data sheet, a self-contained exercise document, states that for level C, undercut depth must not exceed 0.1 times the parent metal thickness with a maximum of 1.0 mm, and for level B the limit is 0.05 times thickness with a maximum of 0.5 mm. Measurement shows a continuous undercut, coded 5011, of 0.8 mm depth over 60 mm length. A plausible mistake in this case is rejecting the weld because you recalled the level B cap of 0.5 mm, or accepting it casually because 0.8 mm is under 1.0 mm without writing any calculation.
The better decision runs the chain: code 5011, level C per specification, limit equals 0.1 times 12 mm equals 1.2 mm, capped at 1.0 mm, so 0.8 mm is within the level C limit. Disposition: acceptable; record location, depth, and length in the report. Why it matters: the same physical flaw can be acceptable at one level and rejectable at another, so the specified level is the anchor of the decision, and showing the calculation is what makes the decision auditable. Repeat this pattern with level B and level D data on the same flaw until switching levels feels routine rather than novel.
- Always quote the quality level from the specification, never from memory of a similar job.
- Write the limit calculation with the thickness before comparing the measurement.
- A complete disposition states code, measurement, limit, decision, and record location.
Scenario 2: separating welder approval from procedure approval
Welder qualification and procedure qualification are independent checks with independent ranges, and a production weld needs both. Verifying one approval establishes nothing about the other, which is what makes document-review cases conceptually distinct from measurement cases.
Paper case: production is running to a WPS that calls for a 24 mm butt weld. The WPS references a WPQR. The assigned welder holds a valid certificate to the ISO 9606 family for the relevant process and position. A plausible mistake in this case is signing off the paperwork on the strength of the welder's certificate alone, reasoning that an approved welder makes the job compliant. The overlooked check is whether the WPQR actually covers the joint: your case data sheet states the WPQR qualified a thickness range of 8 to 20 mm.
The better decision performs two traces. First, procedure support: the WPS variables must fall inside the WPQR ranges, and 24 mm does not, so the WPS is not supported for this joint and production should stop pending resolution. Second, personnel: the welder's qualification range covers process, position, and material group for the job. Only when both traces pass is the production document set complete. This scenario matters because it teaches that approval documents answer different questions: the WPQR says the procedure can produce acceptable welds within its ranges, the welder certificate says the person can execute approved procedures within their own ranges, and neither substitutes for the other.
- Trace the WPS to the WPQR and check each essential variable range, including thickness.
- Check the welder certificate range separately against the actual joint and position.
- Treat a failed trace as a stop-and-escalate condition, not a note for later.
A case-analysis exercise with a self-check rubric
Build a repeating five-step case loop and score yourself on it. The rubric, not a feeling of familiarity, tells you when an imperfection family is genuinely learned and ready to be mixed with others.
The exercise: write or collect five paper weld cases. Each case needs a described joint with thickness, a specified quality level, one or more described imperfections with sizes, and a small data sheet stating the limits you may use. For each case, complete the chain: name and code each imperfection per ISO 6520, quote the quality level from the specification, calculate the applicable limit from the data sheet, decide acceptable or not acceptable, and write a one-line report entry with location, size, and disposition. Do one imperfection family at a time for the first rounds, for example five undercut cases, then five porosity cases, before mixing.
Expected observation: in early rounds you will catch yourself reaching for a remembered limit instead of reading the data sheet; that is the habit the exercise exists to break. By around the tenth case, reading the data sheet first should feel automatic and your reports should read consistently. Self-check rubric, one point each out of five: correct code and name; quality level quoted from the specification; limit calculation shown with thickness; decision consistent with your own calculation; report entry includes location, size, and disposition. Treat a sustained score of four or more as a learning milestone that clears you to move to a new imperfection family. These scores measure practice progress only and are not a prediction of any exam result.
- Five cases per round; one imperfection family until the rubric is consistently met.
- Rubric: code, quoted level, shown calculation, consistent decision, complete report line.
- Milestone of four out of five clears a family for mixing, not for predicting exam outcomes.
An adaptable preparation sequence and concrete readiness checks
Sequence your preparation as decision-skill phases rather than page-by-page reading. Finish each phase with an observable readiness check so you know the skill exists before the next phase stacks onto it.
A realistic sequence you can compress or stretch: Phase one, build the standards-to-questions mapping from the table earlier in this guide by hand, adding a sentence per row on what the standard delivers. Phase two, imperfection drilling: classification and coding until ten common imperfections come out as name plus code plus location plus size without hesitation. Phase three, acceptance cases using the Scenario 1 pattern across quality levels. Phase four, document-review cases using the Scenario 2 pattern, tracing WPS to WPQR and welder ranges. Phase five, mixed timed cases where you must identify which decision type each question belongs to before answering.
Readiness checks before you consider the decision-skill work complete: you can classify ten common imperfections with codes unaided; you can compute an acceptance limit from a given data sheet without re-reading the rules; you can trace a WPS to a WPQR and state which variable ranges you checked; you can write a compliant one-line inspection report from a described flaw; and you can state, for your tier, when you would refer a finding upward instead of deciding. Administrative matters such as entry routes, schedules, and fees for the IWI credential are set by IIW and the Approved National Bodies that administer it, so confirm those details with the issuer rather than any third-party summary.
- Phase order: mapping, classification, acceptance cases, document traces, mixed timed cases.
- Each phase ends with an observable check, not a feeling of coverage.
- Confirm credential logistics only with IIW and its Approved National Bodies.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.
