§3.2A · Under NDE methods
Magnetic particle or penetrant: picking the method on the CWI
Choosing between magnetic particle testing (MT) and penetrant testing (PT) starts with the metal. A yoke needs steel it can magnetize, and penetrant needs a surface the dye can get into and climb back out of. After that, ask how far below the surface the flaw you're worried about could sit.
- Exam
- CWI
- Clause
- §3.2A
- Tags here
- 6
- Parent
- NDE methods
3.2A.1MT and PT on one sheet
| Point | MT | PT |
|---|---|---|
| What the part must be | Ferromagnetic: carbon and low-alloy steel, ferritic grades | Any nonporous material, magnetic or not |
| How deep it sees | Surface and slightly below | Open to the surface only |
| What makes the mark | Particles pile up where a flaw leaks the magnetic field | Dye seeps back out of an opening into the developer |
| Flaw direction | Shows best across the field, so you check twice | Doesn't matter if the flaw is open |
| Field kit | Yoke and dry powder, or wet particles | Cleaner, penetrant and developer cans |
| AWS endorsement scope | Dry powder, yoke technique | Type II visible dye, Method C solvent removal |
≠ marks a row where the two differ.
Endorsement scope per AWS (AWS CWI page, checked October 2026). How those exams work is on the CWI endorsements page.
3.2A.2Start with the metal, then look at the surface
Put a magnet on it. If it doesn't stick, MT is off the table. Aluminum, copper alloys and titanium all leave you with PT for surface work.
On plain carbon steel you usually get to choose, and MT tends to win on a job site. It's quicker, it doesn't need long waits, and it can pick up a tight crack sitting just under a thin skin of sound metal. That reach is shallow, though. A slag line buried in the middle of a thick groove weld is a job for radiography or ultrasonics.
What the surface does to each method
PT lives or dies on the flaw's mouth being open. Grinding, wire brushing or blasting can smear metal over a crack and seal it, and then no dye gets in. Cleaning has to take out oil, scale and paint without closing anything up.
MT is less fussy about a thin coating, but heavy spatter, loose scale and sharp changes in section all collect powder. A steep weld toe can hold a line of particles with no crack under it. That's a nonrelevant indication, and you need to know one when you see it.
Temperature matters too. Penetrant materials have a working temperature range written into the procedure, so very hot pipe or frosty steel in winter needs a procedure that covers it.
3.2A.3Penetrant terms you'll see in a procedure
- Dwell time
- How long the penetrant stays on before the excess comes off. The written procedure sets it. Cutting it short on a tight crack is how indications get missed.
- Excess removal (solvent-removable)
- Wipe with a dry cloth, then with a cloth dampened with solvent. Spraying solvent straight onto the part can flush penetrant out of the flaw you're looking for.
- Developer
- A thin, even white coat that draws penetrant back out and gives contrast. Lay it on too thick and small indications drown.
- Bleed-out
- The spread of dye from an indication. A sharp line points to a crack. A round spot points to a pore.
3.2A.4An MT check with a yoke, start to finish
Clean and dry the area
No grease, loose scale or heavy spatter where the powder needs to move freely.
Set the yoke legs
Place them so the field runs across the direction you expect cracks. A flaw lying along the field lines barely disturbs it.
Dust on the powder with the yoke energized
Apply it lightly, then blow off the excess gently while the field is still on.
Read it before you let go
Look at the pattern with the field on. Particles bunched in a sharp line deserve a second look.
Turn the yoke about 90 degrees and repeat
Two shots at right angles mean a flaw in any direction crosses the field at least once.
Record, clean and demagnetize if needed
Leftover magnetism can make the arc wander on a later repair weld, so follow the procedure on demag.
3.2A.5Magnet or dye?
Material first, flaw location second. Work through them in that order.
0 of 6 tagged · 0 accepted
Tag 01
Which NDE method is most effective for detecting surface-breaking defects in non-magnetic materials?
Remarks on every option
- A RT is volumetric and weak on tight surface cracks.
- B ET does find surface cracks in conductive metals, but PT is the general method for surface-breaking defects in nonmagnetic welds.
- C Correct: PT works on any nonporous material, magnetic or not, and finds discontinuities open to the surface.
- D MT needs a ferromagnetic material. It won't work on nonmagnetic metals.
Pick an option. The remarks on all 4 open here.
Tag 02
In magnetic particle testing, what type of defects are best detected?
Remarks on every option
- A MT only reaches slightly below the surface. Deep internal flaws need UT or RT.
- B MT picks up near-surface flaws, not voids at any depth.
- C Correct: MT finds surface and slightly subsurface discontinuities, and only in ferromagnetic materials.
- D MT needs a ferromagnetic part. Nonmagnetic materials go to PT.
Pick an option. The remarks on all 4 open here.
Tag 03
What is the primary limitation of liquid penetrant testing?
Remarks on every option
- A PT is one of the cheapest methods. Cost isn't its limitation.
- B Correct: penetrant only finds discontinuities open to the surface. Anything below the surface is missed.
- C PT needs only spray cans or a simple station. No power or complex gear.
- D PT results are read on the spot after developing, not sent off to a lab.
Pick an option. The remarks on all 4 open here.
Tag 04
What is the primary limitation of using magnetic particle testing on austenitic stainless steel welds?
Remarks on every option
- A Correct: austenitic stainless isn't ferromagnetic, so you can't magnetize it and MT doesn't work.
- B Cost isn't the issue. On austenitic stainless, MT doesn't work at all.
- C Surface finish affects MT on any steel. The deal-breaker here is the material.
- D Speed isn't the limitation. A nonmagnetic material can't be tested by MT.
Pick an option. The remarks on all 4 open here.
Tag 05
When should penetrant testing NOT be used?
Remarks on every option
- A PT works fine on curved surfaces. Coverage and removal just take care.
- B Cold parts slow penetration and need a longer dwell or warming. It's a limit, not a ban.
- C Bright light is fine for visible-dye PT. Darkness matters for fluorescent.
- D Correct: on porous materials, penetrant soaks in everywhere, so the background bleeds and hides real indications.
Pick an option. The remarks on all 4 open here.
Tag 06
What is the correct sequence for liquid penetrant testing?
Remarks on every option
- A Develop and remove are swapped, and the dwell is in the wrong place. Penetrant needs its dwell before removal.
- B Developer before penetrant makes no sense. The penetrant has to go in first.
- C Correct: clean, apply penetrant, dwell, remove the excess, apply developer, then examine.
- D You clean before applying penetrant, not after. Develop and remove are also out of order.
Pick an option. The remarks on all 4 open here.
3.2A.6Sources
- AWS CWI page — MT and PT endorsement scope, checked October 2026
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.