Army Seeks to Detect Failure Before it Happens

Researchers are developing ways to see into the future.

Maintenance of equipment is a huge expense and requires machine downtime. There’s simply no substitute for good inspection, however, especially in critical applications such as aviation. Army researchers hope to take the time element out of inspection and actually detect material failures that haven’t even happened yet.

As described in an Army News article, the concept is damage precursor detection. Essentially, the goal is to detect changes that precede failure. In many situations the precursor to failure is significant deformation, which is quite noticeable. Unfortunately, there are many other situations where no clear indication is given prior to development of damage, such as a crack. This can allow catastrophic failure to occur between inspections, even if the inspections are done regularly.

The trouble-maker is fatigue. The cyclic loading of components can cause them to fail, even if that load is well below their normal yield point. A crack is obviously a bad thing, and is often a precursor to complete failure. Inspection looks for damage, such as a crack, but it would be even better if the crack could be predicted instead of observed.

Developing methods to detect damage precursors is not a simple task. Other sensors, such as strain gauges, may still determine the structure is “healthy.” Ed Habtour and Dan Cole paired up to combine expertise to study damage using system dynamics and micromechanics in order to seek out the earliest signs of failure.

Theoretical calculations were missing the true behavior, and the model had to be modified. Habtour describes the update, “The missing piece in the dynamic model was the fatigue-damage-precursor component that was captured through micromechanics and detected by adding precursor to the nonlinear dynamic model.”

This addition meant there was something traditional mechnical testing was missing about the microstructural state of the material. The researchers analyzed high stress areas of components using nanoindetation and found that there was localized softening of the structure in these areas.

Cole describes the finding, “The indentation elastic modulus dropped by approximately 40 percent with respect the rest of the structure, which was a huge surprise.” After significant testing and retesting, the trend was confirmed.

While it is only in the beginning of development, the concept of detecting impending failure is one worth pursuing. The method of continuously collecting data on damage precursors is still in the works and the extension to other materials such as composites is not yet known, but the work is a solid step forward in the effort to improve technology for the Army-led Future Vertical Lift Program, which aims to replace current aircraft, such as helicopters , with more advanced versions.


Image: US Army