Dual Acoustic Gravitational Wave Detector
Gravitational waves are perturbations of gravitational field propagating at light-speed . They were predicted in 1916 by Einstein’s Theory of General Relativity and their effect first had been indirectly observed by Hulse and Taylor  along 10 years of radio emission measurement of the binary pulsar PSR 1913+16 from 1974. They discovered a systematic shift in the observed time of periastron relative to that expected if the orbital separation remained constant in the aforementioned binary. Recently, a new binary pulsar was discovered in 2003, the PSR J0737- 3039  and the predictions for energy loss due to gravitational waves appear to match the theory. GWs are the weakest longrange physical signals of nature. Compared with the analog electromagnetic waves power emitted by an electron oscillator, there is a factor 10-53 that decreases the corresponding GW power . That is the reason of the difficulties found to detect them and the 50 years delay for the detector development. A big effort was made in the last 40 years to build GW resonant acoustic detectors , and only in the last few years they were improved to reach a theoretical sensitivity able to detect intra-galactic GW burst sources like supernovae. New generation GW detectors are being designed to achieve deeper sensitivity for weaker GW sources emitting into a few kHz region. They are broadband non-resonant detectors.
To achieve a sufficient sensitivity for 100 kPc distant sources (embracing the Virgo Cluster), dual has to be equipped with a wide bandwidth leverage amplifier. Multi-objective optimization was used to design the highest gaining device vs. the highest resonance frequency of fundamental normal mode of the amplifier.
To keep a broadband bandwidth, the transducer has not to be resonant. So we need a readout system able to accomplish out of resonance measurements. We designed a kind of “speed amplifier” which is a compliant mechanism and does not affect the dynamic of dual bodies. This device is called “leverage” because it works like a lever (its gain depends on the geometrical features of its shape) and is machined starting from a monolithic piece. We used the ANSYS environment and optimized two different amplifiers: a three and four-joints leverage. The main feature of these devices is their geometrical constant gain factor, which does not dependent on the working frequencies. Because of its fundamental vibrational mode must be much higher than the highest frequency of dual bandwidth, our amplifier can be studied as if it were a low frequency device (static gain gives the right value of amplification for the full working bandwidth). We used ModeFRONTIER  software to improve the four-joint leverage amplifier (see fig. 4). Now we are studying how to incorporate the amplifier inside the facing areas of dual. The first prototype was machined in Alumold 600 and tested: we measured its mechanical transfer function and now we are building the apparatus to perform a first thermal noise measurement in order to establish its experimental behavior.