Fred i was asked this question and wanted to know what your thoughts were on this. R and D asked me what was the criteria to decide if to test at a component level or at a system level , my answer was that it should depend on what is the reliability and confidence level of the component
Good question – and it’s not only a factor of reliability and confidence. While those are important to have in mind prior to designing a life test, it’s not the only consideration.
Often the decision to test at the component level is because it has a unique or new failure mechanism which is possible to evaluate and characterize with life testing directly on the component or test structure with the component. It’s often less expensive, easier to accelerate and to accomplish.
The testing at the system level is more expensive, more difficult to focus the testing on a specific component or component failure mechanism. Yet, is often the only way to evaluate interactions between elements of a product or while the product is operating.
In the rare case when a single component and it’s failure mechanism dominant the system’s failures, then testing the component at the system level makes sense.
Ok, backing up a little to your question.
Test at the component level when you want to learn about the life of the component and the components specific failure mechanisms.
Test at the system level when you are exploring system life during expected use conditions (possible to accelerate – like daily temperature changes affect on product life). Or, when the failure mechanisms are related to component interactions and operation.
If you do not have a clear failure mechanism as the target for the testing, then it may be difficult to design the appropriate test. Using FMEA, HALT or some other discovery method is useful to uncover the product’s failure mechanisms. Then move into life testing at the appropriate level with the test focused on specific failure mechanisms.
Hope that helps, please do let me know if you have any questions.