Technology Advances Help Researchers Gain New Insights into Structural Health

Materials used in aerospace vehicles, offshore assemblies, civil structures, and consumer products can all show degraded performance due to fatigue. Real-time prognostics of such structures can help logistics, increase consumer confidence, and improve safety. In fatigue loading, damage generally leads to visible crack initiation and growth which can eventually result in catastrophic failure. Many health monitoring techniques in the literature rely on sensing techniques that capture crack formation and crack growth throughout a structure’s lifetime. However, CALCE researchers are developing a novel class of early-warning fiber optic sensors to detect fatigue damage precursors, i.e. distributed microstructural damage/evolution that precedes fatigue cracking.

To achieve this, a fiber optic conjugate stress sensor (FOCSS) has been developed to track degradation in the effective stiffness of materials undergoing fatigue softening. Conjugate stress sensing relies on two co-located extensometers of significantly different stiffness. Thus far, the extensometers used have consisted of a compliant optical fiber Fabry-Perot interferometer mounted on a compliant plastic pad and a stiff fiber Bragg grating sensor mounted on a stiff steel pad. Analysis of the stress transfer mechanisms provides an appropriate transfer function to estimate the stiffness of the host structure from the strains in each extensometer. Depending on the configuration and geometry, the transfer function has been computed using three different methods: a detailed numerical 3D finite element (FE) method, an analytic 3D approximation using Eshelby’s equivalent-inclusion method, and a simple 1D spring model. The transfer function map obtained from the simple spring-model is compared to the detailed FE model in Figure 1 for the current sensor configuration being fabricated in this project. The sensitivity of the sensor is seen to increase greatly as the host compliance drops. The region of maximum sensitivity can be tailored by increasing the stiffness of the stiff extensometer.

In initial monotonic tensile tests, we first verify the ability of the FOCCS to provide accurate stiffness estimates. Figure 2 shows the stress-strain curve of copper 110 estimated from the FOCSS (using the 1D spring model), compared to the base-line result obtained from a conventional mechanical test system (MTS). Considering the minimalism of the FOCSS and spring model, the FOCSS does a good job of predicting the changes in the tangent stiffness with the onset of plasticity in the host.

The current effort with the FOCSS is focused on capturing fatigue in metals and composites exposed to vibratory environments. It is known that the elastic modulus of materials will change with cyclic loading, either due to grain reorientation in metals or matrix microcracking in composites. Presently, the challenge with this approach is decreasing the noise of the FOCSS to detect small changes of stiffness in a noisy vibratory environment. Therefore, the presented FOCSS capable of real-time, in-situ measurement of fatigue precursors can be of great benefit to many industries.

For more information on this research, contact Prof. Abhijit Dasgupta.

Published February 10, 2020