When it comes to making optical fiber for use in sensors that measure vibration and acoustic pressure, determining wave speed and mechanical loss can mean the difference between a sensor that works and one that does not. A team at Georgia Institute of Technology in Atlanta has designed a method for measuring these properties accurately and noninvasively, using a laser Doppler vibrometer to detect sound waves as they travel through a segment of fiber.Using a laser Doppler vibrometer, researchers at Georgia Institute of Technology have developed an accurate and noninvasive means of measuring wave speed and mechanical loss in optical fiber for sensing applications.Brad M. Beadle, a researcher on the project, explained that astronomers employ this method to determine whether a star is approaching or receding. "We apply the same idea when we use the laser Doppler vibrometer to measure the surface velocity of the fiber," he said. "The scale's just a little smaller than the stellar case."A function generator, power amplifier and ultrasonic transducer generate sound at 75, 100, 125 and 200 kHz, which is coupled into the fiber with a horn amplifier and which travels through it in the form of longitudinal waves. Light from an argon-ion laser emitting at 514 nm passes through a Bragg cell, polarization-maintaining fibers and a series of gradient-index cylindrical lenses to the surface of the fiber, where an avalanche photodiode captures the reflected laser light.The voltage signal from the photodiode is proportional to the velocity at which the sound waves drive the optical fiber. By measuring the temporal phase shift and amplitude attenuation of the elastic sound pulse as it propagates between several locations on the fiber, the researchers deduce the speed of the wave and the loss factor of the fiber.'Intelligent' sensor designBecause the technique is nondestructive and noninvasive, it may be suitable for use on the factory floor to determine the integrity of optical fibers as they are produced. But it also may enable researchers to tailor-make better sensors."It is critical to know the dynamic properties of the fiber in order to intelligently design fiber optic sensors," Beadle said. For example, by knowing the speed of sound in the fiber and the mechanical loss factor, it should be possible to construct a sensor that responds maximally to surface displacement in a particular bandwidth. He said that, because they have information about the dynamic properties of the fiber, they could quickly redesign the sensor if a different sensing bandwidth were desired.