A returned board that passes a bench test has not necessarily disproved the customer’s observation. The test may omit the triggering condition, or handling may have changed an intermittent connection. Record what was attempted and what remains different before closing the investigation.
TI’s failure-analysis information emphasizes device history, usage, failure characteristics, and prior analytical findings in a return. Those inputs are equally useful when preparing a board-level reproduction plan, even when the eventual investigation involves several component suppliers.
Preserve the reported event
Capture the symptom, observed frequency, operating sequence, connected equipment, software version, supply conditions, and any environmental context available. Distinguish direct observations from the customer’s proposed cause. “Communications stopped after wakeup” is evidence; “the processor is defective” is an interpretation to investigate.
Record the board’s received condition before cleaning, reseating connectors, updating software, or reworking it. Those actions can change the evidence. If an action is necessary to perform the investigation, log it and its timing explicitly.
Compare field and bench conditions
| Condition | Reproduction question |
|---|---|
| Power and startup sequence | Is the reported transition being exercised? |
| Loads and peripherals | Does the bench represent the actual connected system? |
| Software and configuration | Is the same behavior enabled? |
| Duration and repetition | Has the relevant opportunity for failure occurred? |
Keep a comparison board where useful, but do not let its passing behavior replace testing of the returned unit. Use controlled changes to narrow the difference, retaining a record of which configuration produced each observation. Preserve original log timestamps and event sequences when the symptom depends on an intermittent transition.
State the result narrowly
- Define the attempted reproduction conditions.
- Record the actual observations and logs.
- List known differences from the reported use.
- Identify further evidence that could discriminate the cause.
- Issue a disposition that reflects those limits.
A useful conclusion may be that the symptom was not observed under the listed bench conditions. Avoid expanding that into a claim that the board can never exhibit the fault. Intermittent behavior and incomplete field information limit what a finite test can establish.
If the board is returned to service or retained for further analysis, preserve its configuration and investigation history. Repeated “no fault found” cycles become more informative when each cycle adds specific conditions and evidence rather than repeating the same generic pass statement.
The goal is to narrow uncertainty honestly and make the next investigative step more effective, while keeping the original reported event traceable throughout the process.
Move from reading to a build-ready project
Use the fabrication and assembly checklists to prepare drawings, production files and test requirements.
Open the DFM checklist