It is not enough to focus on NVH, duplicating what you see in the lab or in the vehicle. In an automated plant setting, there are other factors to take into account. In our experience, metric-based decision making for NVH in an end-of-line tester focuses on several main areas:
- Discrimination – Isolating a specific defect signature from others
- Repeatability – A part’s signature provides the same results every time
- Dynamic Range – Grades the severity of the defect clearly
- Distribution – Population provides a normal distribution of values
- Stability – There is no uncompensated change over time
- Understandability – It should relate back to a physical attribute
- Diagnostic Aid – It should suggest a corrective action for the plant
While most of these items are common sense, it is important to recognize that in a plant setting, metrics need to be real and physical. For instance, a repair station operator isn’t helped by the statement “this part’s tonality is too high.” Part of Signal.X’s experience is in applying NVH fundamentals in ways that detect assembly issues on the line while providing insights to operators and engineers alike.
See an example of how we applied this to seat track testing here.
The team at Signal.X represents NVH professionals who understand the fundamentals of NVH signal processing and its application to the production line. Our services and products reflect that experience that is applied to machines that we commission, customers we help, and tools we provide.