Industry Standard Tests vs Real-world Data
Analysis, Data Acquisition, Troubleshooting
Industry standard vibration tests are used by hundreds of labs to test thousands of products every year. But how closely do the standard tests duplicate the real-world operational environment of those products? Is there a valid way to compare standard test fatigue with real-world fatigue?
Vibration Research decided to give it a try.
We placed an ObserVR1000 and multiple external batteries, surrounded by a cushion of packing material, into a common cardboard shipping box. One tri-axial accelerometer was attached to ObserVR1000’s top surface and another was attached to an inside wall of the box. The ObserVR1000 was turned on, the box was closed, and the package was shipped out via a commercial carrier, recording vibrations during its travel.
A simple comparison could have been done using the Power Spectral Density, which shows the base levels of vibration experienced by a product over a period of time. However, since the goal was a Lifetime testing comparison, it was better to use a random test that creates the equivalent to a product’s total life damage.
This type of lifetime test and comparison can be done with damage analysis using a Fatigue Damage Spectrum (FDS).
The recorded data from two road travel segments was used within VibrationVIEW to create and graph an FDS, which we then compared with an FDS generated from the ISTA 3 standard test profile. The FDS graphs for the data recording generally resemble the FDS graph for the ISTA 3A standard test, with some relative under-testing and over-testing.
We continued by using a combined recordings FDS to create a test profile and comparing that with the ISTA 3 profile. Again, largely similar with some areas of significant difference.
This experiment and analysis is not a comprehensive study of packaging behavior, nor is it an evaluation of the ISTA 3 standard. A much larger set of sample data, collected under more controlled conditions, would be needed to accomplish either of those purposes.
However, this case study does demonstrate that important insights can be gained by an FDS comparison between real world data recordings and a test standard. With more data, rigorously collected, the comparisons would be clearer, providing either confidence in the standard or impetus to move on from using the standard.