§3.3 · Judging the weld
Destructive testing for the CWI
Destructive testing of welds cuts, bends, pulls, breaks or etches a test coupon to show what visual and NDE can only suggest. On the CWI you match each test to the question it answers and to the qualification it feeds.
- Exam
- CWI
- Clause
- §3.3
- Tags here
- 15
- Part A minimum
- at least 3%
3.3.1Tests and where they turn up
| Test | Where you meet it | Who calls for it | Edition watch |
|---|---|---|---|
| Tensile (reduced section) | Procedure qualification (PQR) | The code's qualification clause | Specimen count and size: code-specific |
| All-weld-metal tensile | Filler metal classification, some PQRs | The filler metal spec, or the code where it asks | Required only where called for |
| Guided bend (face, root, side) | PQR and welder tests | The code's qualification clause | Bend radius and acceptance: code-specific |
| Fillet weld break / macroetch | Welder and procedure fillet tests | The code's qualification clause | D1.1:2025 updated macroetch acceptance |
| Nick break | Pipeline welder and procedure tests | API 1104 | API 1104 22nd edition on Part C |
| Charpy V-notch (CVN) | PQRs on jobs that need it | The contract or code, only where specified | Read the job's requirement |
| Hardness survey | Procedure work on hardenable steels | The contract or owner's spec | Limits come from the contract |
Code editions per the AWS Exam References and Editions Guide, August 2025, checked October 2026; the macroetch change per the AWS D1.1:2025 summary of changes.
3.3.2Strength first, then the rest
Small minimums, at least 3% of Part A for Destructive Testing Fundamentals and at least 8% of Part B for Destructive Testing and Properties (AWS B5.1:2025, Table 8.1, checked October 2026), but the vocabulary spills into qualification and metallurgy questions.
Keep the strength terms apart. Tensile strength is the peak load per area before the specimen breaks; yield is where it starts to stretch for good; elongation is how far it stretched. The 70 in E7018 is a minimum tensile strength of 70 ksi, a strength figure that says nothing about how the deposit behaves at a notch.
Toughness and hardness belong to the steel and its heat history, so they're taught with heat control and metallurgy and hot vs cold cracking. Here the point is narrower: a steel can pass a tensile test and still fail a requirement the contract adds.
On Part C the same tests come back as code lookups: which specimens a given qualification needs and what makes one fail. Those answers sit in D1.1 Clause 6 in both 2020 and 2025, and the 2025 edition changed the macroetch acceptance, so an answer remembered from an older book can be wrong on the newer one.
3.3.3How each part asks it
Part A asks purpose: which test answers which question about a joint. Part B hands you results and a specification and asks whether they pass. Part C asks what your code requires for a given qualification. Learn the purpose cold, then practice reading a requirement out of a document you've got open. On Part B this area reads like a lab report: values on one side, the specification on the other, and you decide whether the coupon passes.
3.3.4From test coupon to signed record
Weld the coupon to a WPS
The coupon is welded to the written procedure, with the actual settings recorded as it goes.
Inspect it visually
A coupon that fails visual doesn't move on to cutting.
Cut specimens where the code says
Location, number and type of specimens follow the code's qualification clause, which is D1.1 Clause 6 in both 2020 and 2025.
Test and judge
Each specimen is judged against the acceptance criteria in that clause for the edition in force.
Record and certify
Results go on the PQR for a procedure or on the test record for a welder. How those two records differ is on the WPS vs PQR page.
3.3.5Specimens on the bench
Bend, break, pull and etch, one specimen at a time.
0 of 15 tagged · 0 accepted
Tag 01
What is the primary purpose of a guided bend test?
Remarks on every option
- A Hardness has its own test (Brinell, Rockwell, Vickers). A bend test doesn't measure it.
- B Correct: the guided bend stretches the weld surface. Open discontinuities show whether the weld is sound and ductile.
- C Tensile strength comes from a tensile test, not a bend.
- D Porosity may open up on a bend, but the test exists to check ductility and soundness, not to find pores.
Pick an option. The remarks on all 4 open here.
Tag 02
In a root bend test, which surface of the weld is placed in tension?
Remarks on every option
- A The HAZ is in the bend zone, but the surface in tension is named by side: face, root or side.
- B Correct: in a root bend, the root surface is on the outside (convex) of the bend, so it's stretched.
- C That's a side bend, which uses a cross-section of the weld thickness.
- D That's a face bend. The face goes on the convex side.
Pick an option. The remarks on all 4 open here.
Tag 03
When performing a guided bend test, what is the purpose of the mandrel?
Remarks on every option
- A The jig doesn't measure force. The guided bend is pass/fail on the stretched surface.
- B Speed isn't the point. The plunger sets the bend geometry.
- C Correct: the mandrel (plunger) diameter sets the bend radius, and that sets how much the surface is stretched.
- D Rollers or lubrication reduce friction. The mandrel defines the bend.
Pick an option. The remarks on all 4 open here.
Tag 04
What is the maximum size of discontinuities allowed in most bend test specimens?
Remarks on every option
- A Larger than the open-discontinuity limit in common bend acceptance. A defect this size would reject the specimen.
- B Smaller than the usual single-discontinuity limit. A specimen with an open 1/16 in flaw is normally acceptable.
- C Twice the usual limit. That would pass specimens the codes reject.
- D Correct: 1/8 in is the long-standing limit for a single open discontinuity on the convex face in both D1.1 and ASME IX. Check the exact rules in your code.
Pick an option. The remarks on all 4 open here.
Tag 05
During a Charpy V-notch impact test, what is being measured?
Remarks on every option
- A Correct: the Charpy V-notch measures the energy absorbed in breaking a notched specimen. That's impact toughness.
- B Hardness has its own indentation tests. A Charpy test doesn't measure it.
- C Yield point comes from a slow tensile pull, not an impact swing.
- D Tensile strength comes from a tensile test. Charpy measures the energy absorbed in breaking the specimen.
Pick an option. The remarks on all 4 open here.
Tag 06
Which of the following welding joint designs demonstrates the highest ductility based on its percent elongation and reduction of area?
Exhibit
Joint Design Percent Elongation (%) Reduction of Area (%) Single bevel butt joint 12 30 Double V butt joint 15 35 Fillet weld 5 20 Square butt joint 10 25 Remarks on every option
- A The fillet weld has the lowest elongation (5%) and reduction of area (20%) in the table, so it's the least ductile.
- B The square butt joint (10% and 25%) beats the fillet weld but trails both bevel-type joints.
- C Correct: The double-V butt joint has the highest elongation (15%) and reduction of area (35%), the two standard measures of ductility.
- D The single-bevel butt joint (12% and 30%) comes second. The double-V joint is higher on both measures.
Pick an option. The remarks on all 4 open here.
Tag 07
What should be the temperature of test specimens for standard Charpy V-notch testing?
Remarks on every option
- A Many specs call for Charpy tests below room temperature. You test where the spec says.
- B Preheat is a welding control. It has nothing to do with the impact test temperature.
- C Nobody tests near the melting point. Charpy tests are run at or below service temperature.
- D Correct: specimens are brought to the temperature named in the governing spec or code and held there before breaking.
Pick an option. The remarks on all 4 open here.
Tag 08
In a fillet-weld break test, what is the primary characteristic being evaluated?
Remarks on every option
- A Hardness has its own test. The fillet break is about the root.
- B Correct: breaking the fillet open shows whether the weld fused to the root and both members, and whether it penetrated.
- C The fillet break isn't a strength measurement. It's a visual check of the fracture.
- D Corrosion resistance takes environmental tests. Breaking a fillet doesn't show it.
Pick an option. The remarks on all 4 open here.
Tag 09
What is the primary purpose of hardness testing in welding inspection?
Remarks on every option
- A Fusion shows on a macroetch or a bend, not on a hardness reading.
- B Ductility comes from bend or elongation results. Hardness only hints at it.
- C Correct: hardness readings across the weld and HAZ flag hard zones, which are prone to brittle and hydrogen cracking.
- D Penetration shows on a macroetch. A hardness test doesn't show it.
Pick an option. The remarks on all 4 open here.
Tag 10
What type of destructive test is most appropriate for evaluating the cross-sectional penetration of a weld?
Remarks on every option
- A Correct: a macroetch on a polished, etched cross-section shows root penetration, fusion and weld profile.
- B A tensile test gives strength, not a view of how deep the weld went.
- C An impact test gives toughness, not penetration.
- D A bend test shows ductility and soundness, not the shape of the penetration.
Pick an option. The remarks on all 4 open here.
Tag 11
What surface preparation is required before performing a macro etch test?
Remarks on every option
- A Cleaning alone won't show anything. The section has to be ground smooth and etched.
- B Degreasing doesn't reveal the weld. You need a smooth surface and an etchant.
- C Grinding is only part of it. Without etching, the weld zones won't show.
- D Correct: grind or polish the cut section smooth, then etch it so the weld metal, fusion line and HAZ show up.
Pick an option. The remarks on all 4 open here.
Tag 12
How should specimens be sectioned for macro examination?
Remarks on every option
- A A cut along the weld runs down its length. It won't show the penetration or profile across it.
- B Cutting along the fusion line destroys the very boundary you want to see.
- C Correct: section across the weld, perpendicular to its axis. That shows penetration, fusion and profile.
- D An angled cut distorts the dimensions. Macro sections are cut square across the weld.
Pick an option. The remarks on all 4 open here.
Tag 13
What should be evaluated when examining a nick break test specimen?
Remarks on every option
- A Nick-break specimens are broken, not bent to an angle. Bend angle belongs to bend tests.
- B Hardness isn't read on a nick-break. You look at the broken surface.
- C The nick forces fracture through the weld. Strength isn't the result.
- D Correct: break it through the nick and examine the fracture surface for porosity, slag, lack of fusion and other discontinuities.
Pick an option. The remarks on all 4 open here.
Tag 14
Which type of bend test is most suitable for evaluating complete joint penetration groove welds in thick plates?
Remarks on every option
- A Face bends on thick plate are hard to bend properly and only test the face. Side bends replace them.
- B A free bend is unguided and less controlled. Code qualification uses guided bends.
- C Root bends only check the root. On thick plate, side bends replace face and root bends.
- D Correct: side bends test the whole weld thickness through its cross-section. That's why they're used on thicker CJP groove welds.
Pick an option. The remarks on all 4 open here.
Tag 15
Which destructive test is best suited for evaluating the strength of the weld metal and heat-affected zone?
Remarks on every option
- A Nick-break checks the fracture surface for internal discontinuities. It gives no strength number.
- B A free bend checks ductility, not strength.
- C A macroetch shows the profile and penetration of the cross-section. It doesn't measure strength.
- D Correct: the transverse tensile pulls across weld, HAZ and base metal. It shows the joint strength and where it breaks.
Pick an option. The remarks on all 4 open here.
3.3.6Specimens, books and which part asks
Does the CWI expect you to run destructive tests yourself?
You're expected to know what each test is for and to judge results against the code. Certifying welder qualifications is among a CWI's functions (AWS QC1:2016-AMD1, clause 4, checked October 2026), and the specimens behind them are destructive tests.
How much of the detail has to be memorized?
Part A is closed book (AWS Part A information sheet, March 2024, checked October 2026), so the purpose of each test and the limits your study material flags have to be in your head. Specimen layouts and code-specific details can wait for Part C, which is open book with the code you chose.
Where do destructive tests show up on Part B?
Under Destructive Testing and Properties. Part B is open book with the Book of Specifications that AWS supplies (AWS CWI Examination User Guide, September 2026, checked October 2026). More on the session under Part B practical.
3.3.7Sources
- AWS B5.1:2025, Table 8.1 — minimum percentage of questions per domain; checked October 2026
- AWS Exam References and Editions Guide, August 2025 — code editions accepted for Part C; checked October 2026
Ten questions on a break
Our CWI prep app runs on iPhone and iPad. A few questions between jobs beats none at all.