Fiber optic based sensing systems for dynamic displacement and vibration monitoring

The main goal of our thesis is to develop a fiber optic sensor, more precisely a vibration sensor. The thesis consists on developing a working prototype using two different technologies, with two different optical circuits: the first uses two facing fibers, one in transmission and one in reflection, the second instead is based on a Bragg gratings, therefore the fiber is sensitive to contraction/expansion. The first arrangement is rawer and it is based on Plastic Optic Fibers (POF), it is cheaper and easier under both designing and manufacturing aspects. The second one, instead, exploits glass single mode fibers: it has a cost that is orders of magnitudes larger and requires a lot more effort to manage the optical components, because they must be fiber coupled. Both optical circuits have as their output a light signal, which converges to a transimpedance and amplification circuit (signal conditioning), composed by 4 stages. The current provided by the photodiode carries the information about the sensed displacement inside its intensity and, since voltages are easier to manage, it has to be converted through a transamplifier. Then, two different inverting amplifiers are exploited to reach a suitable dynamics and, at the same time, to not saturate the OP-AMPs. The last stage is a buffer to increase the output impedance. The output of this amplifier is read by an acquisition board based on ATMega328 architecture, which communicates through a PC. From what concern the communication between the microcontroller (the ADC board) and the calculator, we decided to implement two different types of communication, serial and wireless. The first one relies on an USB wire that connects the sensor to the PC, while the second one exploits two radio modules working around 2.4 GHz. The receiving radio module is connected to the PC, where a Python multi process code is running. With the first process, the software is capable to catch all the transmitted samples, while the second one is used to generate a live plot that shows the past 500 samples, in order to monitor in real time the external perturbations affecting the sensor. For this thesis we have simulated, designed and assembled, entirely both optical circuits, designed and manufactured the PCBs, we designed all circuits, using a CAD software. For prototyping we decided to mill a PCB using a CNC machine, because it is more robust than a circuit on a breadboard and it will avoid some disturbances and parasitic capacitance. We programmed the microcontroller to sample the input signal and the protocols to connect with the PC via serial and radio communication, and developed a python software for the live plot of the sampled data. Furthermore, the system has been created and tested, confirming its functioning.