Broad band acoustic detectors J.P. Zendri* on behalf of the AURIGA collaboration www.auriga.lnl.infn.it 5 th International LISA Symposium 12-15 July 2004
Acoustic detector noise budget:one dimensional lumped model Quantum mechanics constrain T n 2 2 2 k T ≥ h ≡ ω ⋅ ω B n S ( ) S ( ) FF XX ω 2 4 The noise stiffness K n is unconstrained = ω ω K S ( )/ S ( ) n FF XX 5 th International LISA Symposium 12-15 July 2004
I α ≡ = Transducer Efficency X General result: − 2 2 ω = α × ω ω ≈ α × ω S ( ) S ( ) S ( ) S ( ) FF VV XX II ∝ ω ⋅ ω ω ⋅ ω T S ( ) S ( ) S ( ) S ( ) n FF X X V V II = ω ω α ω ω 2 K S ( )/ S ( ) S ( )/ S ( ) n FF II VV II 5 th International LISA Symposium 12-15 July 2004
Practical limits to the K n Passive transducers 2 ∝ ω ω K E S ( )/ S ( ) For instance capacitive n bias VV II ↓ ↓ − Ω 2 10 MV m / 10 Breakdown Active transducers For instance optical ∝ × K Finesse LaserPower n ↓ ↓ K n OK but problems ÷ ≥ 5 6 10 10 Watt for cryogenics 5 th International LISA Symposium 12-15 July 2004
Methods to increase the efficiency: resonant transducers Mechanical signal amplified before the transduction One mode transducer H.J. Paik, Jour. Appl. Phys. 47 , 1168 (1976) ω = ω Required bar Tr Increment of a around resonance = M / M BAR TR = ν M / M Bandwidth upper limit bar TR BAR Two mode transducer J.P. Richard, Phys. Rev. Let. 52 , 165 (1984) ω = ω = ω Required bar First Last = M / M Increment of a around resonance BAR LAST = ν M / M Bandwidth upper limit bar First BAR Velocity transformer D. Blair et al, Appl. Phys. 20 , 162 (1987) Broad Band amplification but still a complete thermal noise analysis is missing 5 th International LISA Symposium 12-15 July 2004
Auriga method: Resonating matching line ( ) ω I ω i ME 1 ( ) α ω = = 2 M Bias ≡ − ( ) L L 1 ω + ω − ω + ωω 2 2 x ( L L ) L i / Q eff P + L ( L L ) s in eff el el el P in S 1 ω ⇒ ω ω ≡ el el mech L C eff ω L Enhancement of the transducer efficiency ≡ el eff Q el R around the mechanical modes 5 th International LISA Symposium 12-15 July 2004
Tuning the electrical mode to the mechanical modes require a very high electrical Q quality factor This because the equivalent electric mode mass is light 2 C E = M Tr Bias 20 gr . el ω 2 el = Transducer capacitance C Tr Auriga operate with a Q factor of ~ 500.000 5 th International LISA Symposium 12-15 July 2004
Auriga Output PSD at 4.5 K HV Plate 1E-3 Resonances Spurious 3 Main Modes 1/2 ] 1E-4 1/2 [ Φ 0 /Hz 1E-5 S ΦΦ Wide Band Noise Level 1E-6 800 850 900 950 1000 1050 1100 Frequency [Hz] Around the sensitivity bandwidth Auriga noise is dominated by LC thermal noise 5 th International LISA Symposium 12-15 July 2004
Auriga readout Double SQUID white band noise 1.8 450 k T 1.6 = N B n 400 h ω h 1.4 350 2 /Hz] 1.2 300 N � PSD [ µΦ o 1.0 250 0.8 200 − = Auriga Old N 5000 0.6 h 150 0.4 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Temperature [K] 5 th International LISA Symposium 12-15 July 2004
AURIGA present Sensitivity 1E-19 Auriga Now T=4.5 K -1/2 ] Auriga Old 1/2 [Hz T=0.1 K 1E-20 Noise curve prediction at 4.5K S hh 1E-21 800 820 840 860 880 900 920 940 960 980 1000 Frequency [Hz] 5 th International LISA Symposium 12-15 July 2004
Increase the bandwidth: The Dual detector Solution One: Wrong! Dominated by thermal and BA of the reference mass Solution Two: M 2 ~ M 1 but CM 2 CM 1 Solution Three: M 2 ~ M 1 and same CM Dual Detector 5 th International LISA Symposium 12-15 July 2004
Considered geometries for Dual Dual Sphere M.Cerdonio et. al., Phys. Rev. Let. 87 , 031101 (2001) Dual Torus M.Bonaldi et. al., Phys Rev. D62 , 102004 (2003) 5 th International LISA Symposium 12-15 July 2004
Quantum Limited sensitivity curve (dual torus) = ⋅ 1 1 = ⋅ 11 K 1 .0 1 0 N / m K 1.8 10 N m / − − n Mo n Si C ≥ ⋅ ≥ ⋅ 8 8 Q T / 2 10 Q T / 2 10 = = = = = = r 0.25 m r 0.26 m r 0.47 m r 0.82 m r 0.83 m r 1. 44 m − − − − 1 2 int 2 ext 1 2 in t 2 ext = = = = h 2.35 m Tot wei gth 16. 4 t h 3 m Tot weigt h 6 5 1. t In order to get K n ~10 11 N/m a lever geometrical amplification of at least 10 is required 5 th International LISA Symposium 12-15 July 2004
Broad band lever amplification H.J. Paik et al, Proc. Of the first Amaldi Meeting , edited by E. Coccia et al, World scientific Singapore, p201 (1995) Y Auriga design: X Hinges Mirror Mechanical amplification=Y/X=1/ a>>1 5 th International LISA Symposium 12-15 July 2004
Prototype under test:Amplification Material Al7075 Mode Domain Lever Domain Parameters for simulation •T=300 K •Gaussian Beam waste 0.375 mm •Mirrors Fused silica Geometrical gain factor=10 5 th International LISA Symposium 12-15 July 2004
Prototype test:assembling procedure Harmonic oscillator Amplifier 5 th International LISA Symposium 12-15 July 2004
Prototype thermal noise estimation 1E-15 Resonator alone 1/2 ] xx (f) [m/Hz 1E-16 1/2 S 1E-17 1E-18 0 250 500 750 1000 1250 1500 1750 2000 2250 Frequency[Hz] ≈ × Therm Therm S Gain S − − − xx Amplifier out xx Resonator The amplifier thermal noise contribution is negligable 5 th International LISA Symposium 12-15 July 2004
Prototype thermal noise measurement set-up Room Temperature Thermal Noises Measurement: 10 µK Stabilized Thermal Box! NTC Thermistors Technology ア5μK Active ア1mK Thermal Box 1 μ Yag LASER 5 th International LISA Symposium 12-15 July 2004
A possible implementation in Dual-Torus 5 th International LISA Symposium 12-15 July 2004
Dual Torus Therm. and BA noise reduction: Wide area sensing Signal=mean displacement over the readout area=Green surf.+Yellow surf. GW sensitive (quadrupolar) mode: Unperturbed torus Mode Shape Some GW not sensitive mode which provide only noise: Rejected Not Rejected Some of the low order modes are averaged out High order modes are averaged out 5 th International LISA Symposium 12-15 July 2004
Dual Torus Therm. and BA noise reduction: Selective readout = X X meas-Standard Quradrupolar mode (Signal) → X Maximized Meas-Selective Octupolar (Noise) → X 0 Meas-Selective = − + − X X X X X Meas-Selective 1 2 3 4 5 th International LISA Symposium 12-15 July 2004
Example of Thermal and BA noise reduction using selective readout = r 0.25 cm 1 = r 0.26 cm Material:Molybdenum: Dimensions: − 2 int = r 0.47 cm − 2 ext Strain Noise Power spectrum Red:Selective readout Blue: Not Selective readout 5 th International LISA Symposium 12-15 July 2004
Conclusions •Well designed impedance matching lines between the main resonator and the amplifier has been used to widen the useful band of the Auriga detector more than one order of magnitude • Broad band detectors should be obtained using lever transducers. In particular the proposed detector “dual” promises a sensitivity of the order of 10 -23 Hz -1/2 in the frequency range between 1kHz to 5 kHz. 5 th International LISA Symposium 12-15 July 2004
Recommend
More recommend