A Practical Guide — Writing NDE Acceptance Criteria That Link Indication Size to Component Criticality, Not Just a Standard Name
A practical guide to specifying nondestructive examination (NDE) acceptance criteria on a drawing or purchase specification — how to define indication size limits, reference the correct applicable standard, and link acceptance levels to actual component criticality rather than defaulting to a generic callout. Shivam Forge, Rajkot, India. Call +91-9265772827.
A surprising number of purchase drawings call out an NDE method — ultrasonic testing, magnetic particle inspection, radiography — and a standard name, and stop there, treating the standard reference itself as though it were the acceptance criterion. It isn't. Most NDE standards (ASTM E164, ASTM E1444, ASTM E94, and similar) define how to perform and document an examination; they intentionally do not define what size or type of indication is acceptable, because that judgment depends entirely on the specific component's service condition, stress level, and consequence of failure — information only the component's designer holds. A forging destined for a low-stress structural bracket and a forging destined for a rotating aerospace component can be made from the identical material and inspected with the identical method, yet warrant entirely different indication acceptance limits, because the consequence of an undetected flaw differs by orders of magnitude between the two applications. Writing usable NDE acceptance criteria means three things: naming the correct base standard for the method and material, specifying quantitative indication size and type limits (not just "no relevant indications" without a size threshold), and linking the chosen acceptance level explicitly to why the component's criticality warrants it — often by referencing an established acceptance class (such as ASTM E125 casting reference or the acceptance grades within ASME Section VIII) rather than inventing a bespoke limit. A specification lacking any one of these three elements leaves the acceptance decision effectively undefined, which becomes a genuine problem the first time an inspector finds a borderline indication and has nothing concrete to check it against.
Cite the specific standard governing how the examination is performed and how indications are characterized — for example ASTM E1444 (magnetic particle), ASTM E165/E1417 (liquid penetrant), ASTM E94/E1032 (radiography), or ASTM A388/A745 (ultrasonic) — matched to the method and material form actually being inspected.
State the actual maximum acceptable indication dimension — for example a maximum linear indication length in millimeters, a maximum rounded indication diameter, or a maximum indication density per unit area — rather than a qualitative phrase like "no significant indications," which an inspector cannot apply consistently without a number attached.
Where possible, reference an existing acceptance class or severity level from a recognized standard (such as the severity levels within ASTM E125 or the acceptance criteria tables within ASME Section VIII, Division 1, Appendix 4/8) rather than drafting a bespoke numeric limit from scratch, since established classes have already been calibrated against known service performance.
Specify whether acceptance limits differ by location on the component — many drawings correctly apply tighter limits to a critical bore, fillet radius, or high-stress zone than to a non-critical boss or flange face, and that zone-specific differentiation needs to be stated explicitly on the drawing rather than left to inspector judgment.
Before selecting numeric limits, determine what actually happens if a flaw of a given size goes undetected in this specific component — fatigue crack initiation in a rotating part, brittle fracture in a pressure boundary, or simply cosmetic concern in a non-structural bracket — since this consequence assessment is what should drive how tight the acceptance limit needs to be.
A component subject to high cyclic loading or operating near a material's fatigue limit generally warrants a tighter acceptance class than a statically loaded, low-stress component of similar size and material, because fatigue crack growth is far more sensitive to small initial flaw size than static strength is.
Where a component's design safety factor is already conservative relative to expected service loads, acceptance criteria can reasonably be less restrictive than for a component designed closer to its material's allowable stress limits — this coordination between design margin and inspection stringency avoids over-specifying NDE requirements that add cost without a corresponding reliability benefit.
Where practical, note briefly on the specification or in supporting documentation why a particular acceptance class was selected for this component, which helps a supplier's quality team apply judgment correctly on borderline calls and helps future designers understand whether a criterion can be safely reused for a similar but not identical component.
Nondestructive examination acceptance criteria are among the most commonly under-specified elements on forging purchase drawings, and the gap usually isn't a lack of NDE method callout — most drawings correctly name a method (ultrasonic, magnetic particle, liquid penetrant, or radiographic testing) and often a governing standard. The gap is what comes after: a genuine acceptance criterion requires stating what size and type of indication is actually acceptable, and a surprising number of specifications stop at the method and standard name, implicitly treating the standard itself as though it defined acceptance limits. It does not. Standards like ASTM E1444, E165, E94, or A388 define examination procedure, sensitivity requirements, and how indications should be characterized and reported — they deliberately leave the acceptance decision to the purchaser, because that decision depends on information the standard's authors cannot know: the specific component's service consequence.
This is precisely why the same base material, forged to the same geometry and inspected with the identical method, can legitimately carry very different acceptance criteria depending on what the component actually does in service. A forged bracket carrying modest static load in a non-critical location can reasonably tolerate indications that would be entirely unacceptable in a rotating shaft or pressure-retaining component operating under cyclic stress near a material's fatigue limit, because the consequence of an undetected flaw differs by orders of magnitude between the two applications — fatigue crack growth is disproportionately sensitive to small initial flaw size in a way static strength calculations often are not. Acceptance criteria that ignore this distinction, either by being uniformly loose (risking undetected flaws in critical zones) or uniformly tight (adding inspection cost and potentially rejecting acceptable parts in non-critical zones), both represent a specification failure, just in opposite directions.
The practical fix is straightforward but requires the effort of stating it explicitly rather than leaving it implied: name the correct base examination standard for the method and material form involved, state a genuine quantitative indication size and type limit rather than a qualitative phrase like "no relevant indications" that an inspector cannot apply consistently without a number, and — where an established acceptance class already exists in a recognized standard, such as the severity levels within ASTM E125 or the acceptance tables within ASME Section VIII Appendix 4/8 — reference that class rather than drafting a bespoke limit from scratch, since calibrated classes carry a known relationship to service performance that an invented limit does not. Where criticality genuinely varies across a single component's geometry, zone-specific acceptance limits, clearly marked on the drawing, are worth the extra specification effort rather than defaulting to a single uniform limit across the whole part.
For engineering teams and procurement specifiers preparing forging drawings and NDE specifications, Shivam Forge's quality engineering team can review component criticality and application requirements and help confirm or recommend appropriate acceptance criteria before the drawing goes to production. Contact our engineering team at +91-9265772827 or sales@shivamforge.com with your drawing and application requirement for a manufacturability review and quotation.
No. ASTM E1444 and similar method standards define how to perform magnetic particle inspection and how to characterize indications — they do not define what indication size is acceptable for your specific component. You still need to add a quantitative acceptance criterion, either your own numeric limit or a reference to an established acceptance class, or the acceptance decision is effectively left undefined.
It depends on the consequence of an undetected flaw in your specific application. A rotating, fatigue-loaded, or pressure-boundary component generally warrants tighter acceptance limits than a statically loaded, non-critical structural component, because the failure mode and consequence differ substantially even when the material and method are identical.
Yes, and it's often the better approach. Established acceptance classes within standards like ASTM E125 or ASME Section VIII Appendix 4/8 have already been calibrated against known service performance, and referencing an appropriate existing class is generally more defensible and easier for a supplier's quality team to apply consistently than a bespoke numeric limit drafted without that calibration history.
Not necessarily. Many components correctly apply tighter acceptance limits to specific critical zones — a bore, fillet radius, or high-stress feature — than to non-critical areas like a boss or flange face. If your component has this kind of criticality variation, it should be stated explicitly on the drawing rather than left to inspector interpretation.
Yes. Our quality engineering team can review a component's service application, stress profile, and existing standard callouts and advise on appropriate acceptance criteria, or confirm whether criteria already specified on your drawing are complete and internally consistent.
Why Choose Shivam Forge
Shivam Forge delivers precision hot-forged components from our integrated Shapar, Rajkot facility — covering forging, CNC machining, heat treatment, and quality inspection under one roof.
