The optical readout for AURIGA

this research is in collaboration with 


The motivation - The working principle - The expected sensitivity - A room temperature implementation - A photo gallery


The motivation

Top of the page

In the effort of improving the sensitivity and widen the bandwidth of the detector, the AURIGA group is carrying out a R&D project to develop a new readout chain, based on laser techniques.  The sensitivity and the useful bandwidth (see below) are expected to improve by more than one order of magnitude with respect to present day figures. This would have the double effect of increasing the chances of signal detection and of allowing a better reconstruction of the signal properties (namely, arrival time and frequency components).


The working principle

Top of the page

The optical readout makes use of a high-Q resonant transducer coupled and tuned to the bar, as present days AURIGA. The difference lies in the way the mechanical signal of the bar vibration is converted into an electromagnetic one: here a resonant optical cavity (a Fabry-Perot cavity) is housed between the bar and the transducer. The cavity is made out of two high-reflectivity mirrors facing each other: one of the mirrors is fixed to the bar while the other to the transducer. The bar vibration induces a time varying relative displacement between bar and transducer, thus modulating the length of the cavity, i.e. its optical resonant frequency. The conversion of the mechanical signal is thus achieved. The resonant optical cavity housed between the bar and the transducer is named transducer cavity

Click here for a larger picture showing the full experimental scheme of the optical readout.

By means of a feedback loop, a Nd:YAG (wavelength=1064nm, near IR) laser source is frequency locked to the transducer cavity: this means that the frequency of the laser beam is actively kept equal to the optical resonant frequency of the transducer cavity. Thus by measuring the laser frequency one extract information about the relative displacement of the bar and the transducer, i.e. about the bar vibration.  The measurement is performed by comparison with a frequency reference, provided by a second resonant optical cavity, named reference cavity: again, this is a Fabry-Perot cavity. 

Therefore the laser beam emitted by the source has to be split in two: one beam is sent to the transducer cavity and the other to the reference cavity. The optical resonance of the reference cavity needs also to be guaranteed; it is now not possible to follow the cavity with the laser frequency, as this is already locked to the transducer cavity. Therefore it is now the reference cavity length that has to be corrected: this is done by means of piezoelectric actuators in a feedback-loop that is active only at low frequency (i.e. up to frequencies much lower that the typical detection frequency).

The measurement is performed in a frequency range around bar and transducer modes, i.e. around 900Hz. The reference cavity must be a good standard in this frequency range: the reference cavity displacement noise needs to be as low as 5x10-20m/sqrt[Hz] around 900 Hz. For a 20 cm long cavity this is equivalent to a fractional frequency noise of 3x10-19/sqrt[Hz], better than what ever achieved!

Delivery of the laser beam to the transducer cavity is accomplished by a single-mode polarization maintaining optical fiber. 


The expected sensitivity

Top of the page

Main figures of the optical readout are listed in the table below; for the explanation of technical expressions please refer to the glossary.

Detector thermodynamic temperature 0.1K Bar mechanical Q 5x106
Laser wavelength 1064nm Transducer mechanical Q 5x106
Transducer cavity length 1cm Reference cavity length 20cm
Transducer cavity finesse


Reference cavity finesse 4x104
Laser power input to transducer cavity 2mW Laser power input to reference cavity 5mW
Resonant transducer effective-mass 11 kg Photodiodes conversion coefficient 0.72A/W

Main noise sources are: the thermal noise, the laser power noise (that acts as back-action noise force in the sense that it determines a noise in the radiation pressure on the transducer cavity mirrors), the seismic/mechanical noise and the electronic noise (essentially the photon counting error at the photodiodes).

The detector sensitivity to be expected with the above parameters is shown in figure 1 below. The vertical axis is the usual Shh: this is the usual way to quote the sensitivity of a gravitational wave detector. To better appreciate the detector improvement expected with the optical readout, this curve is to be compared to the present experimental AURIGA sensitivity.

Figure 1 Expected sensitivity of a ultracryogenic bar detector equipped with the opto-mechanical readout


A room temperature implementation

Top of the page

So far we have fully implemented the optical readout on a bar kept at room temperature: thus we have operated a room temperature bar detector equipped with the opto-mechanical readout. This was a major step towards the realization of a cryogenic bar detector with the opto-mechanical readout. See the photo album for a collection of pictures of the experimental apparatus.

For convenience the experimental setup is slightly different from the one designed for the cryogenic operation and above described: this has no effect on the sensitivity. The experimental setup for the room temperature operation is shown in figure 2 below. The laser is frequency stabilized by locking it to the reference cavity (RC). Part of the stabilized beam is sent to the transducer cavity (TC) by means of a single mode polarization maintaining optical fiber, some 10m long. The length of the reference cavity is adjusted (at frequencies lower than a few Hertz) by means of a piezoelectric actuator (PZT) so that two optical resonances of TC and RC overlap: the error signal proportional to the difference in length between the two cavities is acquired as output of the detector, eventually influenced by a passing by gravitational wave. This same signal is the one used to adjust the length of the RC.

Figure 2

Experimental setup for the room temperature operation. Dashed lines mark out active thermal stabilization. The red continuous line shows the laser beam path.

The main difference with respect to the cryogenic operation are:

  • the bar and transducer temperature: about 300K instead of 0.1K;
  • the usage of higher transmittivity mirrors for the transducer cavity (this explains why the finesse is only 2.8x104 instead of the planned 3x105);
  • the usage of a lower mass transducer: this transducer came from a previous design when a lighter transducer was required;
  • the poor quality factor of bar and transducer: this comes from the room temperature operation. In a cryogenic environment the mechanical quality factor of the resonances are expected to raise.

For a better comparison we list below the main figures for the room temperature operation; for the explanation of technical expressions please refer to the glossary.

Detector thermodynamic temperature 296 K Bar mechanical Q 1.8 x 105
Laser wavelength 1064 nm Transducer mechanical Q 6.6 x 103
Transducer cavity length 6 mm Reference cavity length 11cm
Transducer cavity finesse

2.8 x 104

Reference cavity finesse 4.4x104
Laser power input to transducer cavity 1.3 mW Laser power input to reference cavity 5mW
Resonant transducer effective-mass 1.7 kg Photodiodes conversion coefficient 0.72A/W

See also the experimental setup in the photo gallery.

The sensitivity to gravitational waves of this room temperature detector is shown in figure 3 below: the worse sensitivity (ie higher Shh) with respect to the ultracryogenic case is essentially due to the higher thermal noise, caused but the higher temperature and lower mechanical quality factors. Note that the frequency range where the sensitivity is higher (between 850Hz and 900Hz) is different from that expected for the ultracryogenic case: in fact lowering the temperature increases the resonant frequencies of the mechanical oscillators (namely the bar and the transducer) which the sensitivity is peaked to.

Figure 3

Experimental sensitivity of the room temperature detector equipped with the optical readout

For a more thorough discussion of the experimental setup and results see the following paper: L. Conti et al., Room temperature GW bar detector with opto-mechanical readout, Jour. Appl. Phys., 93 (2003) 35, also in gr-qc/0205115.


Previous Next