§0.2 · Timed mock
CWI mock exam, on the clock
Timed, closed book, verdicts held back: the CWI mock exam here runs a shorter Part A set against a clock. Hand in and every tag gets its verdict, with your score split by area. A real Part A form is longer: 165 items, 150 of them scored, in 135 minutes (AWS Part A information sheet, March 2024, checked October 2026).
- Exam
- CWI
- Tags
- 55
- Clock
- 63 min
0.2.1Set it up like the test room
Part A is closed book. Shut the study guide, turn the phone face down, and keep only a calculator of the kind AWS allows: four-function, construction or non-programmable scientific (AWS CWI Examination User Guide, September 2026, checked October 2026).
The spread leans toward the clauses with bigger Part A minimums in AWS B5.1:2025 where the question bank allows, so joining and cutting gets the most tags and destructive testing among the fewest. It isn't a scaled copy of those minimums. Qualification gets no tags at all: B5.1:2025 sets no Part A minimum for it, and it counts on Parts B and C instead. The exam format sheet lays out all three parts.
0.2.2Part A under the clock
Press Start when you're set. You can change a pick as often as you like until you hand in or time runs out.
Before you start the clock
- 55 tags, 63 minutes. The clock starts when you press Start.
- No verdicts while the clock runs. Change a pick as often as you like.
- Hand in early or let the clock run out. Then every tag reads ACCEPT or REJECT, with remarks on each option.
- Your score comes back split by area.
| Area | Tags |
|---|---|
| 1.1 Terms and definitions | 8 |
| 1.2 Welding and NDE symbols | 3 |
| 2.1 Joining and cutting | 13 |
| 2.2 Welding application and control | 7 |
| 2.3 Heat control and metallurgy | 6 |
| 2.4 Welding calculations | 3 |
| 3.1 Visual examination | 3 |
| 3.2 NDE methods | 4 |
| 3.3 Destructive testing | 2 |
| 4.2 Inspector duties | 6 |
0 of 55 tagged
63:00
Tag 01§1.1 Terms and definitions
What does 'GTAW' stand for in welding processes?
Remarks on every option
- A Not a real process name. Nothing in GTAW is "guided"; the word you want is the shielding gas.
- B Made up. "Transfer" belongs to GMAW metal transfer modes, not to the process name.
- C Correct: Gas Tungsten Arc Welding: a nonconsumable tungsten electrode under inert gas shielding. Shop slang is TIG.
- D Close, but the T is tungsten, not transfer. The tungsten electrode is what sets GTAW apart from GMAW.
Pick an option. The remarks on all 4 open here.
Tag 02§2.1 Joining and cutting
Which welding process is most appropriate for achieving maximum penetration in a single pass on thick steel plates?
Remarks on every option
- A GMAW can run deep in spray transfer, but it doesn't match SAW's current or single-pass penetration on thick plate.
- B FCAW deposits fast and penetrates well, but it still runs well below the currents SAW carries on heavy plate.
- C Correct: SAW runs very high current under a blanket of flux, so you get deep penetration and big deposits in one pass on thick plate.
- D SMAW is limited by electrode size and current. It takes several passes on thick plate.
Pick an option. The remarks on all 4 open here.
Tag 03§1.2 Welding and NDE symbols
What does the (PT) symbol indicate in NDE notation?
Remarks on every option
- A Pressure and leak tests have their own letters in the NDE symbol system. PT isn't one of them.
- B Correct: PT is penetrant testing. The method finds discontinuities open to the surface.
- C Coating thickness isn't an NDE method letter on a welding symbol. PT is penetrant.
- D Proof testing has its own designation. PT on an NDE symbol means penetrant.
Pick an option. The remarks on all 4 open here.
Tag 04§3.1 Visual examination
In the welding quality process, which inspection method comes first?
Remarks on every option
- A RT is a volumetric method. It comes after the weld has passed visual, not before.
- B Correct: visual comes first, before, during and after welding. It's cheap, it catches the obvious rejects, and the other methods assume a weld that already passed VT.
- C UT looks for internal flaws. Spend that time on welds that already passed visual.
- D PT is a surface follow-up. Don't run it before the weld has been looked at.
Pick an option. The remarks on all 4 open here.
Tag 05§2.2 Welding application and control
What indicates proper welder performance in controlling distortion?
Remarks on every option
- A Correct: a planned weld sequence and good tacks balance the shrinkage and hold fit-up. That is how a welder controls distortion.
- B More heat input means more shrinkage and more distortion. It's the opposite of control.
- C One long continuous run piles up shrinkage in one direction. Backstep or balanced sequences exist to prevent that.
- D Extra reinforcement adds weld metal that shrinks, plus a stress raiser. It doesn't stiffen anything usefully.
Pick an option. The remarks on all 4 open here.
Tag 06§3.2 NDE methods
Which condition must be met before performing ultrasonic testing on austenitic stainless steel welds?
Remarks on every option
- A The scan surface needs to be clean and smooth enough for coupling, not polished. The grain inside the weld is the problem.
- B Couplant has to transmit sound, but its type doesn't fix scattering from coarse austenitic grains.
- C Cranking the gain raises the grain noise right along with the signal. It doesn't solve the scatter.
- D Correct: coarse austenitic weld grain scatters sound. Lower frequencies (longer wavelengths) get through with less scatter.
Pick an option. The remarks on all 4 open here.
Tag 07§2.3 Heat control and metallurgy
Which microstructural phase is most likely to form in low-alloy steel when cooling rate is moderately slow?
Remarks on every option
- A Cementite is iron carbide. In a slow cool it shows up as part of pearlite, not as the phase that forms.
- B Austenite is the high-temperature phase. It transforms on cooling and isn't what you end up with in low-alloy steel.
- C Martensite needs fast cooling (a quench). A moderately slow cool is too slow to form it.
- D Correct: a moderately slow cool gives austenite time to transform to ferrite plus pearlite.
Pick an option. The remarks on all 4 open here.
Tag 08§2.4 Welding calculations
If electrode consumption is 2.5 lbs/hr and operating factor is 40%, what is the actual electrode usage per 8-hour shift?
Remarks on every option
- A Correct: 2.5 lb/hr × 8 hr × 0.40 = 8 lb. Operating factor is the share of the shift the arc is actually on.
- B That's 2.5 × 8 × 0.5. You used 50% instead of the given 40% operating factor.
- C That's 2.5 × 8 with the operating factor left out. The arc isn't burning the whole shift.
- D That's 2.5 × 2. The 8-hour shift and the 40% factor weren't applied correctly.
Pick an option. The remarks on all 4 open here.
Tag 09§4.2 Inspector duties
How should conflicts between contract specifications and welding codes be handled?
Remarks on every option
- A You don't pick the easier rule. If the two conflict, the conflict gets resolved, not ignored in favor of schedule.
- B Resolving contract-versus-code conflicts isn't the inspector's call to make alone. Write it up and escalate.
- C Correct: write the conflict down and get a ruling from whoever has authority (the Engineer, owner or jurisdiction). The inspector doesn't settle it.
- D A code invoked by the contract isn't advisory. Treating it that way is how rejected work gets accepted.
Pick an option. The remarks on all 4 open here.
Tag 10§3.3 Destructive testing
When should specimens be marked for identification in the destructive testing process?
Remarks on every option
- A By then you may not be able to prove which specimen came from which test piece. Traceability is lost.
- B Too late. Specimens have already been handled and can be mixed up between cutting and testing.
- C Correct: mark each specimen as soon as it's cut from the test piece so traceability to the welder and coupon is never broken.
- D Waiting for final machining leaves the cut blanks unmarked through the shop. Mark them as soon as they're cut.
Pick an option. The remarks on all 4 open here.
0.2.3This mock next to a real Part A form
| Item | Real Part A | This mock |
|---|---|---|
| Length | 165 items: 150 scored, 15 unscored pretest | A shorter set, shown in the briefing |
| Clock | 135 minutes | Scaled down to the shorter set |
| Where | A computer at a Prometric test center | Any browser |
| Book | Closed, nothing on screen | Closed, if you keep it honest |
| Scoring | Correct over total, rounded down to a whole percent; 72% to pass | Percent right by area, no pass line |
| Feedback | Preliminary score emailed by Prometric within two hours | Every tag marked, with remarks on each option |
≠ marks a row where the two differ.
Sources: AWS Part A information sheet, March 2024 and AWS CWI Examination User Guide, September 2026, both checked October 2026.
0.2.4After you hand in
Read the area split first
The lowest area is the next thing to work on. Its clause page in the CWI study guide has tags you haven't seen.
Open the remarks on your misses
A confident miss tells you more than a guess. Read why your option fails, then why the key holds.
Drill it, then come back another day
Run the weak clause through the chips on the CWI practice test. The study plan sets a mock midway through the 12-week runway and another in the last week.
Don't treat the number as a forecast
A mock score here doesn't predict a Part A result, and AWS publishes no pass rates to check it against. Why the real parts feel hard is on the is the CWI exam hard sheet.
Keep it in your pocket
Got an iPhone or iPad? Our CWI prep app keeps practice questions on hand when you're nowhere near a desk.