Bogie frame fatigue test with an unplanned stress exceedance
A vehicle manufacturer submits a new 1,435 mm gauge EMU bogie frame for type approval. The design is claimed to meet EN 13749 category B-VI, and the fatigue test plan specifies 10^7 cycles at a principal stress range of 150 MPa in the side beam. During the witnessed test at an accredited laboratory, the frame survives 10^7 cycles, but four strain gauges on the transverse beam record 172 MPa at the weld toe, above the design value. The laboratory's report states 'test passed, no cracks'. The manufacturer argues that because no crack appeared, the higher measured stress is irrelevant and requests the certificate. The safety authority asks you, as the assessment body, whether the test evidence supports the claimed category, and whether cross-acceptance under the relevant TSI can be granted without further analysis. The bogie is also fitted with a newly designed primary spring that has no service history.
1. Does the fatigue test result support the claimed EN 13749 category B-VI? Set out the assessment steps and the decision you would give.
Read the worked answer
No, not on the evidence as presented. The pass criterion for a category B-VI frame is not merely 'no crack after 10^7 cycles'; the test must demonstrate that the measured stress range at every critical location is at or below the allowable value for the material, weld detail and safety factor of that category. Four gauges reading 172 MPa against a 150 MPa design value is a 15% exceedance, which is outside normal measurement uncertainty if the gauges were calibrated and the amplifier zeroed.
The first step is to check the measurement chain: gauge type and location versus the approved test plan, calibration certificate, temperature compensation and cross-talk. If the readings are valid, the design stress basis is wrong, and the fatigue damage must be recalculated with the real stress spectrum using the appropriate S-N curve and Miner damage sum. A re-test at 172 MPa for 10^7 cycles, or a design modification, would be needed. The laboratory's 'passed' statement does not bind the assessment body.
2. The primary spring has no service history. What additional evidence would you require before granting approval?
Read the worked answer
A new spring design without service history cannot rely on 'proven in use'. I would require: design calculation to EN 13906-1 (or the applicable spring standard) with fatigue and surge analysis; material certificates and hardness; fatigue testing of a representative sample at the load spectrum, including mean stress and dynamic amplitude, with a statistically valid number of specimens; failure mode and effects analysis covering fracture, relaxation and corrosion; and a validation plan showing that the spring's stiffness and height tolerances keep the bogie within its wheel-load transfer limits.
Because the spring affects running safety and track forces, I would also ask for vehicle-level validation (ride and curve negotiation) and a first-fit inspection regime, for example height and re-torque checks at 5,000, 50,000 and 100,000 km. Approval could be conditional on these results, with a configuration hold point before fleet release.
3. Can cross-acceptance under the TSI be granted? Explain the role of the accredited test report.
Read the worked answer
Cross-acceptance is not automatic and the test report is evidence, not a certificate. Under the TSI assessment regime, the assessment body must verify conformity against the relevant requirements, including the bogie structural and running-dynamic requirements and any national rules. An accredited laboratory's report establishes traceability and competence, but it does not relieve the assessment body of judging whether the test configuration, load cases, gauge positions and pass criteria match the claimed category.
Here the report contains an unplanned measurement above the design value, so the assessment body cannot conclude conformity. A positive decision could only follow resolution of the exceedance and assessment of the updated evidence; otherwise a negative assessment or a request for further information is the correct output. Cross-acceptance in another Member State would also require the receiving authority to accept the evidence, and it may impose additional national requirements.