A passed durability test means the sample met the specified requirement under the specified conditions. It does not mean every service condition has been covered. The difficult part of tensioner validation is not running the test longer, but explaining the relationship between the test conditions and the real service risk.
A bare wheel body, a wheel assembly with bearing, and a complete tensioner answer different questions. A wheel assembly may focus on bearing and wheel-face condition; a complete tensioner may also involve tensioning force, damping and swing behaviour, depending on the design.
The test plan should state the sample condition, the mounting method, the associated components and the measuring positions. Results from different objects cannot simply substitute for one another, and a bearing-only test should not be described as a life validation of the complete assembly.
Test planning should be built around the failure modes to be validated — for example a change in lubrication or sealing condition, bearing wear, abnormal wheel-face abrasion, or a change in tensioning mechanism performance. Each risk needs an observable indicator and a defined way of judging it.
Whether temperature, speed, load, swing and environmental factors are applied in combination should follow the design requirement, the actual duty cycle and the purpose of the validation. Simply raising one variable does not automatically make an effective accelerated test; if the dominant failure mechanism changes, the result may no longer explain real service life.
A temperature reading on its own is not enough to judge product condition. The report needs to state sensor position, fixing method, ambient conditions, speed, load, running time and whether the agreed steady state was reached.
Readings from different measuring points cannot be compared at will. If non-contact measurement is used, the influence of the measuring conditions on the reading should also be confirmed. Assess measurement repeatability during test development, then set limits from the product requirement — rather than treating an occasional reading as an anomaly.
Noise from a tensioner may originate in the bearing, the contact surface, the mounting condition or an associated component. The motor, the drive and structural vibration of the bench itself can also enter the measurement.
Noise testing should therefore define speed, load, mounting condition, sensor position and background environment. For an abnormal sample, analyse it together with vibration, rotational condition and teardown findings rather than judging the cause from a single overall sound level.
Recording only the start and end condition can miss an abnormality that appeared and then disappeared. Depending on the purpose, schedule intermediate checks on temperature trend, noise change, rotational resistance and related performance parameters.
The checks themselves can disturb the test — through shutdown cooling, disassembly or relubrication. Check frequency, permitted operations and the recovery rule after an interruption should therefore be written into the plan before the test starts, not decided on the spot.
Reaching the specified running time is only one end condition; it does not replace a judgement of product condition. Before the test, define the permitted performance change, the abnormal-stop conditions, the visual and teardown inspection requirements, and what to do if data is missing.
If a sample fails, record the time, the duty condition and the symptom, and retain the necessary physical evidence. Replacing the failed part, continuing to accumulate hours and then presenting the total as a pass for the original sample is not acceptable.
An equipment enquiry should state:
For temperature, speed and load, distinguish the nominal capability of the machine from what it actually controls and measures under test conditions. Bench acceptance can use representative samples and check repeated mounting, repeated testing, abnormal stops and data recovery. Where a relationship with field behaviour is needed, revise the plan step by step against real failed samples; no single bench test can be promised to predict every market life. The durability and temperature-rise benches described under automotive parts testing and intelligent assembly equipment can serve as a reference for duty range and data output.
Related equipment: the value of a bench lies in controllable conditions and readable data. Control accuracy for temperature, speed and load, and the way curves are exported, are set out in the equipment parameters and acceptance notes on the detail page.
Other process notes in this series:
If the test conditions are not written clearly, a pass only describes the sample on that one day. Fix the duty cycle, the measuring points and the decision rules, and the bench data becomes reusable.
See automotive parts testing & intelligent assembly equipment →