If your plating operation runs hydrogen embrittlement testing under ASTM F519, you are already working within a framework that invokes ASTM E8. When a sustained load test specimen goes to a tensile machine for sample qualification, E8 governs that pull. So when E8 changes — and the August 2025 publication of E8/E8M-25 introduced significant changes including formal definitions of ASTM E8 replacement test vs retest — the ripple reaches your process control operation whether the F519 specification has caught up yet or not.
For the first time, E8/E8M-25 formally defines the difference between a replacement test and a retest. These are not interchangeable terms. They describe fundamentally different situations, carry different documentation requirements, and — critically — they provide a clarity framework that F519 currently lacks but will inevitably adopt. This article walks through what changed, why it matters for your process control operation today, and what you should be doing about it before your next audit.
What Changed in E8/E8M-25
The previous edition, E8/E8M-24, did not include formal definitions for terms the testing industry uses every day. The -25 revision, developed under ASTM Work Item WK81396 and approved July 1, 2025, added defined terms to Section 3.1 and two new procedural sections — Section 7.14 (Invalid Test) and Section 7.15 (Retest) — along with updated reporting requirements in Section 8. The key definitions:
| Term | Definition (paraphrased from Section 3.1) |
|---|---|
| Valid test | A test meeting all requirements in the test method and applicable specifications |
| Invalid test | A test whose results are unrepresentative due to identifiable causes |
| Conforming test result | Result of a valid test that conforms to requirements |
| Non-conforming test result | Result of a valid test that does not conform to requirements |
| Replacement test | A test conducted to replace the results of an invalid test |
| Retest | A test conducted to repeat a valid test that produced a non-conforming result |
| Test record | Document maintained by the testing laboratory containing observations and derived data |
| Test report | Document issued by the testing laboratory describing the test and results |
The Key Distinction: Invalid Test vs. Non-Conforming Result
The entire framework hinges on one question: Was the test valid? A valid test meets all requirements in the specified test method and applicable specifications. It can produce either a conforming or non-conforming result — both are legitimate data. An invalid test is one whose results are unrepresentative due to identifiable causes: equipment malfunction, specimen preparation errors, traceability loss, or procedural deviations. An invalid test tells you something went wrong with the testing process, not the material.
This leads to two distinct paths:
- Replacement Test (Section 7.14): When a test is invalid, a single replacement test may be performed using the same method and sample parameters. The invalid result is not used for specimen disposition, but the reason for invalidity must be documented in the test record.
- Retest (Section 7.15): When a valid test produces a non-conforming result, retesting may be permitted — but only under conditions specified by the governing material or product specification, or by the customer. The original non-conforming result must be maintained in the test record.
A replacement test addresses a problem with the test. A retest addresses a problem with the result. You cannot “re-anything” without first making a decision on the rationale for the failing result. Both E8 and F519 prioritize root cause analysis before proceeding, and this shared principle is what makes the framework defensible.
What This Means for ASTM F519 Process Control
ASTM F519 directly invokes E8 for sample qualification testing — the tensile portion of the hydrogen embrittlement test method. That direct reference means E8/E8M-25’s new definitions are already in play for qualification pulls. But the implications extend beyond the areas where E8 is explicitly invoked.
Same Concept, Not in Conflict, but Less Prescriptive
F519’s use of “replacement testing” aligns conceptually with E8’s replacement of an invalid test — the specimen had a visible shank failure, an equipment anomaly, a machining defect. F519’s references to “retesting” echo E8’s retest of a valid but non-conforming result. The intent is consistent. But F519 is substantially less diligent with this language, blending terms that E8/E8M-25 now carefully separates. A review across F519 editions shows an evolution similar to E8’s own trajectory — incremental refinements to reduce ambiguity, but without the formal definitions E8/E8M-25 has now established.
Is there a conflict between the two standards? No. F519 is more outcome-focused and generally limits its use of “replacement” or “retest” to anomalous sustained load test failures. F519 additionally contains clear deference to cognizant engineering judgment regarding testing protocols, including step-load and time-off-load allowances and waivers. Nothing in F519’s current language contradicts E8’s new framework. The gap is prescriptiveness, not direction.
Harmonization Is Inevitable
Given the absence of conflict, the common objectives among ASTM committees to reduce ambiguity, and the continuous contribution of expertise from interested parties, harmonization between F519 and E8 on this language is not a question of whether but when. E8 is leading the way. Incremental or wholesale adoption of its prescriptive definitions will find its way into all tension-based testing standards and likely beyond. The E8/E8M-25 definitions were developed with the intent of cascading through ASTM E6 (the master terminology standard for Committee E28) to other subcommittee standards.
While we publish here about E8, we recommend applying the same rigor and common vocabulary to any testing following a failing result, regardless of which specification governs the test. If a process control test under F519 produces a failing result, be clear in your internal documentation about the root cause assessment, whether the follow-on testing constitutes a replacement test or a retest in accordance with E8’s definitions, and what that distinction means for your records.
Documentation, Communication, and Audit Readiness
E8/E8M-25 is explicit about what must be recorded. For an invalid test, the reason for invalidity must be maintained in the test record (Section 7.14.3). For a retest, the original non-conforming result must be maintained (Section 7.15.3). During a NadCap or customer audit, the way you describe a repeated test communicates something about your process. “We replaced an invalid test” tells the auditor the testing process had an identifiable issue. “We retested a non-conforming result” tells the auditor you received a valid data point indicating a potential process excursion. Auditors will expect your documentation to support whichever narrative you present.
At Omega, when we observe a visible anomaly — a shank failure, surface corrosion, a machining defect — we proactively communicate that to the customer as part of the testing process. On the rare occasion where a laboratory-caused issue affects a test, we report that transparently. This is root cause documentation in practice: the determination of validity happens before anyone discusses next steps. Your internal records should reflect the same discipline.
What’s Still Developing
Section 7.15 places the authority for defining retest conditions with the governing specification or customer requirement. At the November 2025 ASTM Committee E28 meeting, a concern was raised that the published retest provisions do not explicitly address limits on further retesting when the governing specification is silent. ASTM Work Item WK97154 has been opened to address this. The follow-on work is expected to further refine and expand the framework — clarifying retest parameters rather than restricting the definitional structure. These are furthering modifications, not limiting ones. The core definitions are stable and should be adopted now.
What You Should Do Now
We recommend proactive alignment with E8/E8M-25 vocabulary as a matter of documentation clarity, improved communication with your testing laboratory, and clean records that unambiguously align with the latest adjacent standard from ASTM. Start by ensuring your team understands the distinction. When a process control test fails, the first question is always why — and the answer determines the path forward.
Update your internal documentation practices. When preparing a new sample set following a failing result, be explicit about the root cause assessment and the status of the follow-on test as either a replacement test or a retest per E8’s definitions. This applies to F519 process control testing now, not just where E8 is directly invoked.
Knowing that auditors will be focused on this rollout in the coming year, proactive adoption will at minimum reduce potential misinterpretation and at best avoid a contentious finding in your next audit. Don’t swim upstream — it is inefficient and exhausting. E8/E8M-25 has given the processing and process control segment of our industry an opportunity for clarity where there was ambiguity, and this implementation is better done before it is required.
When a result doesn’t go the way you expected, call your testing laboratory. At Omega, we answer the phone. We can help you determine the right path forward, document the situation correctly, and provide context from over 40 years of aerospace process control testing.
References
- ASTM E8/E8M-25, Standard Test Methods for Tension Testing of Metallic Materials, ASTM International, West Conshohocken, PA, 2025. DOI: 10.1520/E0008_E0008M-25.
- ASTM F519-23, Standard Test Method for Mechanical Hydrogen Embrittlement Evaluation of Plating/Coating Processes and Service Environments, ASTM International.
- ASTM Work Item WK81396 — Original work item producing the E8/E8M-25 definitions revision.
- ASTM Work Item WK97154 — Follow-on revision addressing retest provision refinements.
