Effects of bonding layer on the available strain measuring range of fiber Bragg gratings

Abstract

Stress-induced birefringence can lead to distortion in the reflection spectra of fiber Bragg grating (FBG) sensors, thereby resulting in the loss of accuracy and stability of strain measurements. The bonding layer is a direct factor in producing stress birefringence within FBGs. To assess the impacts quantitatively, a theoretical model that links the bonding layer and the reflection spectrum was established. At the same time, the finite element method, based on the theoretical model, was used to study the relationships between characteristics of the bonding layer and reflection spectrum in detail. The analytical results indicate that high elastic modulus and mismatched Poisson's ratio of bonding layer decrease the available strain measuring range of FBGs remarkably, and that unreasonable geometric parameters of the bonding layer should be avoided. In addition, a validation experiment was conducted and experimental results proved the prediction of the theoretical analysis. It can be concluded from the results that the bonding layer is the major limiting factor for the application of surface-bonded FBG sensors in large strain measurements. The bonding materials and bonding processes used in producing FBG sensors deserve serious consideration.

Response of fiber Bragg gratings bonded on a glass/epoxy laminate subjected to static loadings

“…Surface bonding of FBGs leads to questions about strain transfer. To analyze the strain transfer mechanism from a loaded substrate to the optical sensor, theoretical approaches have been developed to provide prediction tools [11][12][13][14][15][16][17]. Numerical analyses using finite element modeling (FEM) are then proposed to validate the analytical models [11][12][13][14].…”

“…Numerical analyses using finite element modeling (FEM) are then proposed to validate the analytical models [11][12][13][14]. Only a few authors have validated their analytical approaches experimentally [13,15,16]. Ansari and Yuan [15] tested bonded coated fibers as well as bare fibers on an aluminum beam subjected to bending.…”

“…They concluded that when optical fibers are used without any coating (bare fiber) a total strain transfer from substrate to sensor is obtained. Zhang et al [16] compared two adhesive processes, epoxy resin and sprayed metal. FBGs were bonded on a galvanized steel wire.…”

“…In summary several studies suggest that the thickness between the sensor and the substrate should be as small as possible [12,16,17] and that the polymeric coating should be removed [15]. These studies also suggest that the bonded length should be longer than the sensor gauge length [12,16] to avoid misleading longitudinal strain gradients and that the overall bonding layer thickness must be reasonable to prevent distortion of the FBG's reflected spectrum [12,16,17]. Similar distortions may take place if the adhesive is too stiff and bulky [16].…”

“…These studies also suggest that the bonded length should be longer than the sensor gauge length [12,16] to avoid misleading longitudinal strain gradients and that the overall bonding layer thickness must be reasonable to prevent distortion of the FBG's reflected spectrum [12,16,17]. Similar distortions may take place if the adhesive is too stiff and bulky [16]. If the stiffness of the substrate is low compared to that of the optical fiber strain transmission loss is possible [13].…”