CRI exam — how many weld acceptance criteria do you actually need to memorize?
Taking the CRI Certified Radiographic Interpreter exam in about 5 weeks. I've been working in NDT for 4 years, mostly UT, but I moved to a shop that does radiographic interpretation and my supervisor wants me certified within 90 days. The study material is dense.
My main concern is the weld discontinuity acceptance criteria. AWS D1.1, ASME Section V, and API 1104 all have slightly different acceptance limits and the exam reportedly pulls from all three. I keep confusing which standard allows what porosity cluster size or incomplete fusion tolerance. Do I need exact millimeter values or is it more about recognizing the category of defect?
My approach so far: reading the ASNT study guide, reviewing actual radiographs from my shop, and doing about 2 hours/day. Scored 66% on a timed practice set last week and hoping to get to 80%+ before the real thing.
Any advice on where most candidates struggle? Specifically curious about IQI placement and density range questions — those feel abstract compared to just reading a film.
66% is pretty close to where I was 5 weeks out and I passed at 78%. The density range and film density questions clicked once I started doing them under timed conditions rather than looking things up. Speed matters on that exam.
IQI placement questions were a bigger chunk of my exam than I expected. Know the 2% rule cold and understand the difference between hole-type and wire IQIs. I probably saw 8-10 questions that touched on IQI in some way.
The actual film interpretation portion is more forgiving than the code section in my opinion. If you're 4 years in NDT you probably have the visual pattern recognition down. The code knowledge gaps are where most experienced techs get surprised.
For acceptance criteria I'd focus on the differences between standards rather than memorizing every value. They like to test whether you know that ASME and AWS treat linear indications differently. That comparison framework is more useful than rote memorization.
Honestly, looking back, I stressed way too much about memorizing every single acceptance criteria table. What actually showed up on the exam was more about understanding the logic behind the criteria — knowing that porosity limits differ between structural and pressure vessel codes, or why undercut depth matters differently depending on the service environment. The specific numbers you need cold are the ones in AWS D1.1 and ASME Section V, but more importantly you need to recognize which code applies to a given scenario before you even look at the acceptance column.
Four years in UT actually helps more than you'd think. You already understand discontinuity types and their formation mechanisms — that's half the battle. The shift is learning to characterize what you're seeing on film rather than what you're hearing in a waveform. Elongated indications, whether something is planar vs. volumetric, aspect ratios — that's where your experience translates. The parts that tripped up people in my cohort were film density ranges and the geometric unsharpness calculations, which feel more formulaic if you haven't been doing hands-on RT interpretation daily.
Five weeks is tight but doable if you're already working in the field. Focus your last two weeks almost entirely on practice questions — reading the explanation for every wrong answer matters more than re-reading the reference material at that point. The exam has a way of wording scenarios that sounds ambiguous until you've seen enough variations. Repetition is how you build the pattern recognition to work through those quickly under time pressure.
I failed my first attempt last year, and honestly it was because I did exactly what you're describing. I made these huge memorization sheets for every acceptance criteria table and tried to brute force all of them. Bad idea. On the actual exam it's less about reciting exact numbers and more about knowing which code applies, what the process is for looking things up, and being fast at applying the criteria to what you're seeing in the image. Second time around I stopped memorizing tables cold and instead drilled the logic, like porosity vs slag rounded indications, when cumulative length matters, that kind of thing. The numbers stuck naturally once I understood why the limits exist.
The other thing that killed me the first time was image quality questions, which I'd basically ignored. IQI placement, density ranges, artifacts vs real indications. I lost easy points there. I ran through the cri image quality artifacts stuff over and over until it was automatic, and that section went from my worst to my best. With 4 years in NDT and 5 weeks out you're in decent shape. Just don't make it a pure memory game like I did.
I passed the CRI last year and honestly, trying to memorize acceptance criteria as a flat list nearly sank me. What actually worked was doing practice questions and then spending most of my time on the ones I got wrong, figuring out exactly why each wrong answer was wrong. Like, is this indication rejectable because of its length, its type, or the code section I'm working under? Once you can explain that, you've basically learned three or four criteria at once instead of one.
The exam loves answers that are almost right. They'll give you a discontinuity that's acceptable under one condition and rejectable under another, and if you only memorized the "right" answers you'll fall for it every time. Coming from UT you already think in terms of sizing and evaluation, so lean on that. Five weeks is enough if you drill the reasoning, not the flashcards. Good luck.