Requesting reliability test data for fatigue life from stamping parts suppliers sounds simple—until the report arrives. Our sourcing team learned this the hard way on a bracket project in Vietnam.
To request reliable fatigue life data, specify your real service loads, mounting conditions, surface finish, and pass/fail criteria in the RFQ. Then require a documented test plan, at least 28 specimens for reliability claims, statistical analysis, failure evidence, and full traceability to production-state parts.
A vague request gets a vague answer. A supplier who receives “please confirm fatigue life” will send back “passed fatigue test”—and that tells you almost nothing. This guide shows you exactly what to ask for, when to ask for it, and how to check what comes back.
What Fatigue Life Data Should I Require Before Approving a Stamping Supplier?
Last year, a U.S. purchasing manager sent us a "qualified" fatigue report from a previous supplier. It contained one cycle count and one word: passed. Our engineers rejected it in five minutes.
Require the test method and standard used, the loading profile and frequency, sample state and drawing revision, number of specimens, cycle-to-failure or runout results, acceptance criteria, statistical treatment, fracture photos, and traceability to the actual part build. A single cycle count is not reliability data.

Stamped parts like brackets, clips, contacts, and mounts fail under repeated loading in specific ways. They crack, loosen, lose contact force, buzz, or deform. Fatigue life depends heavily on the stress mode, the geometry, the surface condition, and the manufacturing history of the exact part tested. So the data package must describe all of those things, not just a number.
The Minimum Data Package
When we audit stamping suppliers in Vietnam and Taiwan for our clients, we ask for a complete package before design signoff. Here is what it should contain.
| Data Element | Why It Matters |
|---|---|
| Test standard and method | Ensures comparability (e.g., ASTM E466 1, ISO 12107, JIS Z2273) |
| Loading profile, stress ratio R, frequency | Confirms the test matches your real duty cycle |
| Sample state and drawing revision | Prevents prototype data being passed off as production data |
| Number of specimens | Reliability claims need statistical backing, not one part |
| Cycle-to-failure or runout data | Shows actual life, not just a pass/fail stamp |
| Acceptance criteria | Defines what "failure" meant during the test |
| Statistical treatment | Weibull fit, scatter, confidence limits |
| Fracture photos and crack observations | Explains why the part failed or survived |
| Material batch traceability | Links results to chemical composition and mechanical properties |
Screening Data vs. Qualification Data
There is a critical distinction here. Exploratory coupon tests with three or four samples can guide early design direction. But they are not release data. ISO guidance recommends at least 28 specimens when the goal is a reliability claim. Before you approve a supplier for production, insist on qualification-grade evidence built on representative parts—same material batch behavior, same tooling, same finish. In our project files, we tag every fatigue report as either "screening" or "qualification" so nobody confuses the two during PPAP 2 review.
How Do I Ask Suppliers for Reliability Test Reports Without Delaying My Project?
Timing is the trade-off we weigh on every custom parts project. Ask too late, and fatigue testing blocks your launch. Ask too early, and the supplier tests the wrong revision.
Send fatigue requirements with the RFQ, not after quotation. Include drawings, loads, environment, and acceptance criteria upfront, request the supplier's test plan for review before testing starts, and run testing in parallel with tooling. Early alignment prevents retests that cause the real delays.

The biggest schedule killer is not the test itself. It is the retest. When a supplier tests with the wrong fixture, the wrong load direction, or the wrong sample state, you lose weeks. High-cycle fatigue is already slow—reaching 10 million cycles at 20 Hz takes roughly 139 hours per specimen. Multiply that by a proper sample count, and you cannot afford a false start.
The RFQ Checklist That Prevents Retests
Our project managers attach this checklist to every stamping RFQ that carries a fatigue requirement. It front-loads the information the supplier's lab needs before the first run.
- Drawing and revision level.
- Material grade, thickness, temper, coating, and heat treatment.
- Welds, bends, fasteners, and surface finish requirements.
- Service environment and temperature range.
- Load direction, magnitude, and estimated cycle profile.
- Any known vibration standard or customer specification.
- Required sample state (production tooling, production finish).
- Acceptance criteria and failure definition.
- Report deliverables, including raw data files.
Run Testing in Parallel, Not in Series
Here is the workflow we use to protect timelines for our U.S. clients:
| Project Phase | Fatigue Activity | Who Acts |
|---|---|---|
| RFQ | Send loads, environment, criteria | Buyer |
| Quotation | Supplier proposes test plan and fixtures | Supplier |
| Design freeze | Buyer approves test plan in writing | Both |
| Tooling build | Fixtures and instrumentation prepared | Supplier |
| First off-tool parts | Qualification testing begins immediately | Supplier |
| PPAP review | Full report with traceability submitted | Both |
Agree on cycle count, load range, frequency, failure criteria, and fixture details before the first specimen goes on the machine. In our experience managing PPAP submissions from Vietnamese plants, a one-hour test-plan review call saves an average of three to four weeks of downstream schedule risk.
Which Test Standards Should I Reference When Requesting Fatigue Life Data From Asian Suppliers?
A buyer once asked me why two quotes for the same clip cited completely different fatigue methods. The answer: neither supplier had been told which standard to use, so each picked what their lab already ran.
Reference ASTM E466 for force-controlled high-cycle fatigue, ISO 12107 for statistical planning and analysis of fatigue data, and JIS Z2273 when working with Japanese-influenced labs. Match the method to the failure mode: force-controlled for high-cycle, strain-controlled for low-cycle, vibration testing for resonance issues.

Standards do two jobs. They make results repeatable, and they make quotes comparable. Without a named standard, you cannot fairly compare fatigue reports from a Vietnamese supplier and a Taiwanese supplier, because they may define failure, control load, and count cycles differently.
Match the Standard to the Damage Mode
This is the part many buyers skip. The test method must match how your part actually gets damaged in service. Otherwise, the results are misleading no matter how well the test was run.
| Failure Mode in Service | Test Approach | Typical Standard Reference |
|---|---|---|
| High-cycle fatigue (bending, spring action, millions of cycles) | Force-controlled testing, S-N data | ASTM E466 |
| Low-cycle fatigue (large plastic strains, thousands of cycles) | Strain-controlled testing | ASTM E606 methodology |
| Resonance, oscillatory loading, transport vibration | Vibration-based validation on fixtures | Customer or industry vibration spec |
| Statistical planning and reliability analysis | Staircase methods, sample-size guidance | ISO 12107 3 |
| Japanese supply chains, sheet materials | Sheet fatigue methods | JIS Z2273 |
Practical Notes for Asian Sourcing
When we place stamping work in Vietnam, Taiwan, or India for U.S. clients, we specify the standard, the stress ratio R, the frequency, and the waveform in the purchase specification. We also ask which parameters the lab actually controls, because some smaller labs run "reference to" a standard rather than full compliance. Ask suppliers to state any deviations in writing. If your customer spec calls for mean-stress correction 4, name the model—Goodman or Gerber—so the analysis is consistent across suppliers. These details cost nothing to write into an RFQ and eliminate most cross-supplier confusion.
How Can I Verify That My Stamping Supplier's Reliability Test Data Is Accurate and Trustworthy?
Early in our supplier audit work, we found a fatigue report where the tested samples were hand-deburred prototypes, while production parts shipped with sheared edges. The lab work was honest. The data was still useless.
Verify accreditation (ISO 17025 for mechanical fatigue testing), confirm the tested samples match production state, request raw cycle-by-cycle data files, check material batch traceability, review fracture photos against FEA-predicted failure locations, and require data from multiple stamping batches, not one prototype run.

Trust is built from checkable details, not from the report's cover page. Here is how our quality team breaks verification down into layers.
Layer 1: Verify the Lab
Ask whether the testing lab holds ISO 17025 5 accreditation specifically for mechanical fatigue testing. Generic ISO 9001 certification covers the quality system, not the technical competence of the fatigue lab. Third-party accredited results carry independent traceability and impartiality.
Layer 2: Verify the Samples
Fatigue life is extremely sensitive to surface finish, residual stress 6, weld quality, heat treatment, coating, and edge condition. So ask directly: were the tested parts made on production tooling, with production finish, at the current drawing revision? Request material certificates showing chemical composition, tensile strength, and yield strength for the tested batch. If the samples were shot peened, plated, or deburred differently from production, the data does not represent your parts.
Layer 3: Verify the Analysis
Request the digital raw data—cycle-by-cycle load-displacement curves—not just the summary PDF. Your engineering team can then revalidate the statistics independently. Check for Weibull analysis 7, scatter bands, and stated confidence levels such as 90% reliability at 95% confidence. Look for evidence beyond the cycle count: crack initiation observations, stiffness loss trends, and fracture photos. If the supplier ran FEA, ask for the correlation report showing that predicted stress concentrations match the physical failure locations.
Layer 4: Verify Process Variation
One batch proves one batch. Tooling wear and lubrication drift change edge quality over time, which changes fatigue life. Ask for data across multiple stamping batches before treating the result as a standing capability. When our Vietnam team runs final inspections, we periodically pull production parts and compare edge condition against the originally qualified samples—a simple check that catches drift before your customer does.
Conclusion
Vague fatigue requests produce untrustworthy reports, and untrustworthy reports produce field failures. Define your real use case, demand statistics and traceability, verify the lab, and approve only production-representative data.
Frequently Asked Questions
What Fatigue Life Data Should I Require Before Approving a Stamping Supplier?
Last year, a U.S. purchasing manager sent us a "qualified" fatigue report from a previous supplier. It contained one cycle count and one word: passed. Our engineers rejected it in five minutes.
How Do I Ask Suppliers for Reliability Test Reports Without Delaying My Project?
Timing is the trade-off we weigh on every custom parts project. Ask too late, and fatigue testing blocks your launch. Ask too early, and the supplier tests the wrong revision.
Which Test Standards Should I Reference When Requesting Fatigue Life Data From Asian Suppliers?
A buyer once asked me why two quotes for the same clip cited completely different fatigue methods. The answer: neither supplier had been told which standard to use, so each picked what their lab already ran.
Footnotes
1. Provides background on fatigue testing concepts referenced in the standard. ↩︎
2. Explains the part approval process referenced throughout supplier workflow. ↩︎
3. Official ISO body for statistical fatigue data planning standard cited. ↩︎
4. Background on mean-stress correction models like Goodman and Gerber. ↩︎
5. ISO source for lab accreditation standard referenced for verification. ↩︎
6. NIST provides authorit
