11:15 AM - 11:30 AM
[SSS05-18] Validation of Fiber Bragg Grating sensors for strain measurement on giant biaxial rock friction experiments
Keywords:Friction experiments, Strain measurement, FBG sensor
The spatially dense array of strain measurements near the fault is crucial to investigate the nucleation and propagation mechanisms of ruptures in laboratory stick-slip experiments. In general, the strain gauges have been installed along the fault; however, the high-density measurement is impeded by the size of the sensor as well as the wiring complexity. The Fiber Bragg Grating (FBG) sensors have the potential to achieve hundreds of measurement points along the fault by unifying them into an optical fiber, which is used to measure the fault-parallel component of the normal strain εxx along the fiber. In this study, we newly installed the FBG sensors on a giant biaxial rock friction apparatus with a 6-meter-long simulated fault and validated the measurement accuracy by comparing to the strain gauges.
The giant biaxial rock friction apparatus consists of a pair of metagabbro blocks vertically stacked in the outer frame. The interface of the specimens is assumed as the simulated fault. The triaxial semiconductor strain gauges (SKS-30282, Kyowa Electronic Instruments Co., Ltd.) are installed along the fault with a spacing of 260mm each and 10mm above the fault surface. We used them as the comparison basis associated with the εxx. We installed the 44 FBG sensors (Technica Optical Components, LLC.) 25 mm away from the triaxial strain gauges. The FBG sensors with a gauge length of 10mm are imprinted in the single-mode optical fiber. We used the interrogator HYPERION si255 (Luna Innovations Inc.) to record the strain with a sampling frequency of 5kHz. The measurement resolution of the strain is 1.7με, which is estimated from the sensitivity of the FBG sensor and the wavelength accuracy of the interrogator.
We conducted the stick-slip experiments with the normal stress of 2MPa uniformly applied to the rock specimens with a servo-controlled shear loading of 0.01mm/s. We compared the peak amplitudes of the local change in the strain caused by the stress concentration around the rupture front. The strain gauges show the rupture velocity close to the sub-Rayleigh speed with the peak strain change of ~3με. The root-mean-square error of the peak amplitude between the strain gauges and FBG sensors was obtained as 0.74με. Thus, this measurement accuracy of the FBG sensors was high enough to track the rupture propagation. We also compared the strain accumulation during the inter-seismic period of the stick slips. The preslip occurred at the edge of the fault, which was detectable with both the strain gauges and FBG sensors as a moderate change of the trend in the strain time series. Overall, the strain measurement near the fault using the FBG sensor is capable of monitoring the strain changes caused by the evolution of slip on the fault. The enhanced spatial resolution using the FBG multi-channel measurements will help clarify the detailed nucleation process and the propagation of the ruptures in large-scale laboratory experiments.
The giant biaxial rock friction apparatus consists of a pair of metagabbro blocks vertically stacked in the outer frame. The interface of the specimens is assumed as the simulated fault. The triaxial semiconductor strain gauges (SKS-30282, Kyowa Electronic Instruments Co., Ltd.) are installed along the fault with a spacing of 260mm each and 10mm above the fault surface. We used them as the comparison basis associated with the εxx. We installed the 44 FBG sensors (Technica Optical Components, LLC.) 25 mm away from the triaxial strain gauges. The FBG sensors with a gauge length of 10mm are imprinted in the single-mode optical fiber. We used the interrogator HYPERION si255 (Luna Innovations Inc.) to record the strain with a sampling frequency of 5kHz. The measurement resolution of the strain is 1.7με, which is estimated from the sensitivity of the FBG sensor and the wavelength accuracy of the interrogator.
We conducted the stick-slip experiments with the normal stress of 2MPa uniformly applied to the rock specimens with a servo-controlled shear loading of 0.01mm/s. We compared the peak amplitudes of the local change in the strain caused by the stress concentration around the rupture front. The strain gauges show the rupture velocity close to the sub-Rayleigh speed with the peak strain change of ~3με. The root-mean-square error of the peak amplitude between the strain gauges and FBG sensors was obtained as 0.74με. Thus, this measurement accuracy of the FBG sensors was high enough to track the rupture propagation. We also compared the strain accumulation during the inter-seismic period of the stick slips. The preslip occurred at the edge of the fault, which was detectable with both the strain gauges and FBG sensors as a moderate change of the trend in the strain time series. Overall, the strain measurement near the fault using the FBG sensor is capable of monitoring the strain changes caused by the evolution of slip on the fault. The enhanced spatial resolution using the FBG multi-channel measurements will help clarify the detailed nucleation process and the propagation of the ruptures in large-scale laboratory experiments.