a little music a day keeps insanity at bay

aww yisss. optical sensors lab 1, prev. sem.: optical transducer using fiber bragg gratings/recording sound through optical fibers. photo c/o yuting.

Background

Acoustic vibrations are generated by musical instruments either by plucking the strings mounted on an instrument, blowing air into a cavity, and various other means. These vibrations are transmitted through the air and are received by the ear. In some cases amplification and/or recording of these acoustic vibrations is desired and is typically done using piezoelectric transducers that are sensitive to mechanical vibration in the acoustic frequency (from 5Hz to 20kHz). In some cases, magnetic pickups are used which record vibrations using electromagnetic induction. 

A drawback from these methods, however, is that the system involves electrical cables connecting the musical instrument to the recording instrument which generates noise and significant audio signal delays. In order to address some of these issues optical pickups have been developed being potentially insensitive to electrical interference.

Here, we recreate the study performed by Loock, et. Al. on an optical pickup which uses a fiber bragg grating as a sensor. As discussed in Lab 1, an FBG is a periodically modulated structure of the refractive index along the core of a single-mode fiber. Any environmental change that changes the parameters of the FBG leads to a change in the reflection spectrum. By exploiting the sensitivity of an FBG to mechanical strain caused by acoustic vibrations we are able to monitor and transduce those signals.

Experiment 

Figure 1 shows a schematic diagram of the experimental setup. The FBG is pre-strained and fixed on a flat portion of the guitar. The reflection spectrum of the grating is shown in Figure 2 and shows the peak to be at 1520.296 nm. Light signal from a tunable semiconductor laser is guided through an optical circulator to the FBG and is tuned to the steep flank along the long wavelength range of the FBG spectrum.

Fig. 1. Schematic diagram of the experimental set-up. The light from the tunable light source is guided through a cirulator towards the optical pickup (FBG). The third arm of the circulator is connected to a photodiode which monitors the change in transmission.

By doing so, the intensity of the transmitted light would vary depending on how much the FBG peak has shifted. The change in transmission is monitored using a photodiode. The setup is connected to a PC and uses the software Matlab for processing the recorded signal.

Fig. 2. Reflection spectra of the FBG (lower spectrum, yellow) and laser diode (upper spectrum, white).

Results and Discussion 

By slightly adjusting the spectrum of the laser source and changing the gain of the photodiode we can tune the system in order to get a recording with the best signal to noise ratio. Figure 3 shows sample waveforms using different gain values and laser diode wavelength peaks.

Fig. 3. (a)1516.900nm, gain: 50dB, (b)1516.900nm, gain: 40dB, (c) 1516.956nm, gain: 40dB, (d) 1516.960nm, gain: 40dB

Although we can clearly notice the presence of noise in some waveforms, we cannot make a direct association between the recorded waveforms and it’s signal to noise ratio. The quality of the recording is measured by ear and a gain of 40 with a wavelength peak at 1516.956 nm seems to yield the best signal to noise ratio.

The sensitivity of the response depends on the slope of the laser attenuation spectrum where the FBG peak is tuned. It is noticeable that even a very small change in the wavelength settings, in the order of picometers, leads to a considerable change in the quality of the recording. In certain settings we notice a relatively good signal to noise ratio but there exists high harmonics in the recordings (Fig.3a) that may be caused by the FBGs nonlinear response to strain.

The quality of the recordings also depends greatly on the initial strain induced on the fiber as well as the relative distance on which the fiber is affixed on the instrument. When unstrained, acoustic vibrations would not induce large changes in the fiber properties resulting in poor quality recording. This phenomena was observed yet unquantified for this demonstration.

From this we can remark that FBGs can indeed be used as an optical transducer.

::