Development ofCryogenic Torsion-Bar

Gravitational Wave Detector

Satoru Takano, Tomohumi Shimoda, Ching Pin Ooi, Yuki Miyazaki, Yuta Michimura, Masaki Ando

The Univ. of Tokyo24/05/19 GWADW2019 @ Elba

/ 2224. 05. 2019 GWADW2019

Torsion Bar Antenna (TOBA)TOBA : TOrsion-Bar Antenna ・Gravitational wave detector using two torsion pendulums・Resonant frequency of torsion pendulum ~ mHz・→ Sensitive to low frequency ( ~ 0.1Hz)・Target sensitivity h ~ 10-19 / √Hz @ 0.1 Hz with 10 m bars

!2

potential sensitivity and scientific possibility in moderntechnologies.

Principle of a torsion-bar antenna.—A TOBA is com-prised of two bar-shaped test masses, arranged parallel tothe x-y plane and orthogonal to each other (Fig. 1). Eachbar is supported at its center, so as to rotate around thez axis. When GWs pass through this antenna, tidal forcesby the GWs will appear as differential angular changes inthese bars. These changes are extracted as a GW signal byusing a sensitive sensor, such as a laser interferometer.

The rotation angle ! of a test-mass bar from the originalposition is obtained by the equation of motion

I €!þ " _!þ #! ¼ FGWðtÞ; (1)

where FGW is the torque caused by the GWand " and # arethe damping factor and spring constant of the restoringtorque by the support, respectively. The moment of inertiaof the test mass is expressed as I ¼ R

$ðx2 þ y2ÞdV, where$ and V are the density and volume of the test mass,respectively. Assuming that the antenna is much smallerthan the wavelength of target GWs, the torque caused by aGW with an amplitude of hij is expressed as

FGWðtÞ ¼ 14q

ij €hijðtÞ; (2)

by using dynamic quadrupole moment q ij [16]. For barrotation, q 11 ¼ %q 22 ¼ %R

$ð2xyÞdV and q 12 ¼ q 21 ¼R$ðx2 % y2ÞdV.Here, we consider the response of a test-mass bar ar-

ranged along the x direction to GWs traveling along the zaxis, h11 ¼ %h22 ¼ hþ and h12 ¼ %h21 ¼ h&, where hþand h& denote the amplitudes of two independent polar-izations (plus and cross modes, respectively) of incidentGWs. In an approximation that the test-mass bar freelyrotate around the z axis (" ' 1 and # ' 1), Eq. (1) resultsin a simple equation, !1 ¼ %h&=2, where % is a shapefactor of the test mass; % ¼ q 12=I ’ 1 in the case of a thinbar. Another test-mass bar, arranged along the y axis,rotates with an opposite sign as !2 ¼ %%h&=2. The re-sultant output of the antenna is expressed as

!diffðtÞ ( !1 % !2 ¼ %h&ðtÞ: (3)

GWs with a cross polarization are observed as differentialangular fluctuations of the test-mass bars [17].

Now, we consider the situation that the antenna is rotat-ing around the z axis with an angular velocity of!rot. In an

approximation that rotation is sufficiently slow, the re-sponse of the antenna is expressed as

€! diff ¼ %½€h& cosð2!rottÞ þ €hþ sin ð2!rottÞ*; (4)

by calculating the torque, Eq. (2), in a coordinate rotatingwith the antenna. This indicates that the GW signal ismodulated by the rotation; a GW signal with an angularfrequency of !g is up- and down-converted to appear at!g + 2!rot frequencies. Equation (4) results in

!diff ’ %!!g

2!rot

"2½h& cosð2!rottÞ þ hþ sin ð2!rottÞ*; (5)

in the case of !g ' !rot. The low-frequency GW signal isup-converted to signals at an angular frequency of 2!rot.Equation (5) also shows that two polarization componentsof incident GWs are extracted from two quadrature phasesof the antenna output.Advantages of torsion-bar antenna.—A TOBA has sig-

nificant features in both ground-based and space-bornedesigns. As a ground-based antenna, a TOBA is a novelapproach to observe low-frequency GWs. In a usualground-based interferometric antenna, a test-mass mirroris suspended as a pendulum to behave as a free mass in thehorizontal plane. Conversely, it has almost no fundamentalsensitivity to GWs below the resonant frequency of thependulum (around 1 Hz). In a TOBA, a test-mass bar issupported as a torsion pendulum, with a low resonantfrequency on the order of a few millihertz in the rotationaldegree of freedom. Thus, a TOBA has a fundamentalsensitivity to low-frequency GWs.The modulation and up-conversion scheme by antenna

rotation is favorable for the observation of low-frequencyGWs below a few millihertz. Here, we note that the ob-servation run may be an intermittent one; the observationcan be a series of data-taking operations with rotation andreverse rotation. The up-conversion of the GW signal isalso advantageous from a practical perspective. Modu-lation prevents various types of low-frequency noises thatare difficult to suppress, such as drifts of instrumentscaused by daily or seasonal changes in the environmentand 1=f noises of electronics in sensors and controllers.As a space antenna, a TOBA has good compatibility

with spin-stabilized spacecraft. In a spacecraft, spinningitself is a simple and robust way to maintain its attitudewith a gyro effect, without additional disturbances fromattitude controllers. ATOBA, with its rotation axis alignedwith that of the spacecraft spin, has a wide observationband from the low-frequency limit determined by the ob-servation time. Another advantage in a space configurationis that the antenna is free from gravity-gradient and seismicnoises caused by ground motions.Sensitivity limits.—The fundamental sensitivity of a

TOBA is limited by the thermal fluctuation of the bars,readout noise of the angular motion, and effects of the barsupport, as detailed in Refs. [1,18]. We estimate the con-tributions of these noises in the case where a cylindrical baris suspended as a torsion pendulum at its center, and its

FIG. 1 (color online). Principle of a torsion-bar antenna. Twoorthogonal bars feel differential torques by incident gravitationalwaves.

PRL 105, 161101 (2010) P HY S I CA L R EV I EW LE T T E R Sweek ending

15 OCTOBER 2010

161101-2

Design sensitivity of TOBA

Shot noise

Radiation pressure

Suspension thermalBar thermal

Target

M. Ando+ (2010)

/ 2224. 05. 2019 GWADW2019

Science of TOBA

10 Gpc for 105 M⦿

!3

IMBH BinaryMerger

GW StochasticBackground

NewtonianNoise

EarthquakeAlert

Astrophysics

TOBA

Geophysics

baseline length. The advantage of a TOBA is its configu-ration simplicity, the potential sensitivity in low frequen-cies even with a ground-based configuration, and thecapability of an intermittent observation of low-frequencyGWs with a modulation and up-conversion scheme.

A TOBA has an option to be a resonant antenna, inwhich two test-mass bars are connected by a shaft with asmall spring constant [23]. Though the observation band islimited around the resonant frequency in this configuration,the requirement for the angular sensor is relaxed.Moreover, its sensitivity to low-frequency GWs can beenhanced by tuning the resonant frequency to twice theantenna rotation frequency. In such a case, a reduction ofthe thermal noise of the shaft is critical.

Besides the fundamental noises investigated in thiswork, there are many practical noises to be considered:additional noises in the angular sensors, Brownian fluctua-tion by residual gases, and noises due to electromagneticfluctuations [24]. In a ground-based configuration, thesimplicity of a TOBA is helpful for low-frequency isola-tions and common-mode reduction of seismic disturbancesand for the reduction of gravity-gradient noises in anunderground site. A space mission requires a reduction inthe antenna size while maintaining its sensitivity by usingadvanced optical technologies. Optimization of the an-tenna parameters, implementation of advanced interfero-metric techniques, and investigations of these noisebehaviors will be considered in future works.

Conclusion.—We propose a gravitational-wave antenna,a TOBA, comprised of bar-shaped test masses and sensors tomonitor their differential angular motions. This antenna hasa fundamental sensitivity to gravitational waves with fre-quencies lower than 1 Hz, which are inaccessible by currentground-based detectors. In order to investigate the conceptand potential of a TOBA, we are developing a prototypeground-based detector [25]. In addition, we have developeda tiny module, called the SpaceWire Interface Demon-strationModule, for space-related demonstrations; this mod-ule has been operated in a low-Earth orbit for one year [26].

This work was supported by a Grant-in-Aid for YoungScientists (A) and the Grant-in-Aid for the Global COEProgram ‘‘The Next Generation of Physics, Spun fromUniversality and Emergence’’ from the Ministry of

Education, Culture, Sports, Science and Technology(MEXT) of Japan.

*ando@scphys.kyoto-u.ac.jp†Present address: High Energy Accelerator ResearchOrganization, Tsukuba, Ibaraki 305-0801, Japan.

[1] P. R. Saulson, Fundamentals of Interferometric Gravita-tional Wave Detectors (World Scientific, Singapore, 1994).

[2] J. Weber, Phys. Rev. Lett. 22, 1320 (1969); P. Astoneet al., Phys. Rev. D 76, 102001 (2007).

[3] A. Abramovici et al., Science 256, 325 (1992); VIRGOCollaboration, Report No. VIR-TRE-1000-13, 1997;K.Danzmann et al., Max-Planck-Institut fur QuantenoptikReport No. 190, 1994; K.Tsubono, in Gravitational WaveExperiments (World Scientific, Singapore, 1995), pp. 112–114; M. Ando et al., Phys. Rev. Lett. 86, 3950 (2001).

[4] J.W. Armstrong et al., Astrophys. J. 599, 806 (2003).[5] S. E. Thorsett andR. J.Dewey, Phys.Rev.D 53 , 3468 (1996).[6] P. Fritschel, in Proceedings of the SPIE Meeting, Waikoloa,

Hawaii, 2002 (SPIE, Bellingham, WA, 2002), pp. 282–291; F. Acernese et al., J. Opt. A 10, 064009 (2008); K.Kuroda et al., Int. J. Mod. Phys. D 8, 557 (1999).

[7] M. Punturo et al., Classical Quantum Gravity 27, 084007(2010).

[8] LISA System and Technology Study Report No. ESA-SCI11, 2000.

[9] N. Seto, S. Kawamura, and T. Nakamura, Phys. Rev. Lett.87, 221103 (2001); S. Kawamura et al., ClassicalQuantum Gravity 23 , S125 (2006).

[10] S. Phinney et al., NASA Mission Concept Study (2004); J.Crowder and N. J. Cornish, Phys. Rev. D 72, 083005(2005).

[11] S. Dimopoulos et al., Phys. Rev. D 78, 122002 (2008).[12] G. Hobbs et al., Classical Quantum Gravity 27, 084013

(2010).[13] H. C. Chiang et al., Astrophys. J. 711, 1123 (2010).[14] V. B. Braginsky and V. S. Nazarenko, in Proceedings of the

Conference on Experimental Tests of Gravitation Theories(Jet Propulsion Laboratory, Pasadena, 1971), pp. 45–46.

[15] C.W. Misner, K. S. Thorne, and J. A. Wheeler, Gravitation(Freeman, San Francisco, 1973), pp. 1012–1017.

[16] H. Hirakawa et al., J. Phys. Soc. Jpn. 41, 1093 (1976).[17] The antenna directivity for sky positions and polarizations

is similar to that of a conventional laser-interferometricantenna [1].

[18] C.M. Caves, Phys. Rev. D 23 , 1693 (1981).[19] P. Ajith et al., Phys. Rev. D 77, 104017 (2008);

arXiv:0909.2867.[20] R. Takahashi et al., Astrophys. J. 596, L231 (2003).[21] T. Akutsu et al., Phys. Rev. Lett. 101, 101101 (2008).[22] B. P. Abbott et al., Nature (London) 460, 990 (2009);

R. Saito and J. Yokoyama, Phys. Rev. Lett. 102, 161101(2009).

[23] S. Kimura et al., Phys. Lett. 81A, 302 (1981).[24] B. L. Schumaker, Classical Quantum Gravity 20, S239

(2003).[25] K. Ishidoshiro et al., Physica C (Amsterdam) (to be

published).[26] W. Kokuyama et al. (to be published).

102 104 106 108101

102

103

104

Total Mass of Black holes [Mo]

Lum

inos

ityD

ista

nce

[Mpc

]

3 5 8 12 205

SNR

FIG. 4 (color online). Estimated SNR for a merger of equal-mass black holes with equal and parallel spin parameters of 0.5[19], shown as a function of the total mass and the luminositydistance.

PRL 105, 161101 (2010) P HY S I CA L R EV I EW LE T T E R Sweek ending

15 OCTOBER 2010

161101-4

M. Ando+ (2010)

Seismic Newtonian noise(i) Integration is carried out over the seismic field in a

half space. Newtonian noise at the lowest frequen-cies may depend on the Earth’s curvature.

(ii) The field is dominated by fundamental Rayleighwaves. Especially with respect to NN mitigation,one needs to consider possible contributions frombody waves and Rayleigh overtones.

(iii) Effects of underground cavities on NN are ne-glected. Underground detectors in cavities mayalso be sensitive to gravity perturbations from shearwaves [55]. Seismic NN in underground detectorsdepends on the geometry of the cavity, and scat-tered waves contribute to NN. The latter two effectsshould be negligible at low frequencies.

(iv) Rayleigh waves have frequency-independent speed.In reality, Rayleigh waves can show strong disper-sion [56] also below 1 Hz. The speed of continentalRayleigh waves lies within 2–4 km/s between10 mHz and 1 Hz. However, since seismic NN atlow frequencies does not depend significantly onthe speed of seismic waves, implementing a real-istic dispersion should not alter the results verymuch.

(v) Propagation-direction averaged NN is calculatedassuming an isotropic seismic field. It is wellknown that the seismic field can show significantanisotropies especially at low frequencies [57].

Infrasound Newtonian noise(i) Integration is carried out over the infrasound field

in a half-space. The thickness of the atmospherecan be a fraction of the length of infrasound waves.For this reason it should be expected that infrasoundNN is significantly smaller below 0.1 Hz thanreported in this paper. In addition, infrasound wavesare reflected from layers of the atmosphere (i.e. thestratosphere or thermosphere) at characteristic an-gles [54]. Newtonian noise at lowest frequenciesmay depend on the Earth’s curvature.

(ii) Mean air density, air pressure, and speed ofinfrasound waves do not change with altitude.

(iii) The speed of sound is frequency independent. Thereare no studies of the dispersion of atmosphericinfrasound at low frequencies (especially as a func-tion of altitude). For a given infrasound field,dispersion has a weak effect on NN below 1 Hz.

(iv) The atmosphere does not move. Winds play animportant role in the propagation of infrasoundleading to characteristic patterns in the field [54].It is unclear if wind in relation to infrasound waveshas additional consequences for NN apart from thefact that wind can be a local source of infrasoundwhen interacting with surface structure.

(v) Propagation-direction averaged NN is calculated foran isotropic infrasound field. Isotropy is certainly anunrealistic assumption as mentioned before.

Using the seismic spectrum published in [58] and a fit tothe pressure spectrum published in [53], we obtain the NNcurves presented in Fig. 9. As a final remark we want topoint out that both seismic and infrasound NN have lowerlimits since seismic and infrasound spectra both lie aboveglobal low-noise models [59,60]. Therefore, in terms ofsite selection, the goal should be to identify a site whereboth spectra are close to the respective low-noise models.

B. Gravity transients

The GW community has not paid much attention toterrestrial gravity transients in the past except for a paperon anthropogenic noise focusing on surface detectors suchas LIGO or Virgo [61]. The reason for this is that gravitytransients can be eliminated in high-frequency detectorssimply by avoiding abrupt changes in the velocity of mov-ing objects and humans within a zone of about 10 m radiusaround the test masses. The situation is very different forlow-frequency detectors. Even though the terrestrial tran-sient landscape is completely unknown and difficult tomodel in many cases, it is possible to identify potentiallysignificant contributions.a. Newtonian noise from uniformly moving objects.—

We consider the case of an object that is moving at constantspeed v along a straight line that has distance rj to a testmass at the closest approach. Therefore, the vector ~rjpointing from the test mass to the closest point of approachis perpendicular to the velocity ~v. The closest approachoccurs at time tj. As before, we express the result in termsof the Fourier amplitude xjð!Þ of test-mass displacement,

xjð!Þ ¼ 2Gm

v2!ðK1ðrj!=vÞ cos ð!Þ

þ iK0ðrj!=vÞ cos ð"ÞÞei!tj : (5)

Here, m is the mass of the moving object, ! is the anglebetween ~rj and the arm, " is the angle between ~v and the

0.01 0.1 110

−21

10−20

10−19

10−18

10−17

10−16

10−15

10−14

Frequency [Hz]

Str

ain

nois

e [1

/√ H

z]Seismic NNInfrasound NNLaser−AtomTOBAMichelson

FIG. 9 (color online). Seismic Rayleigh wave and atmosphericinfrasound NN together with the sensitivity curves of the threeMANGO concepts.

JAN HARMS et al. PHYSICAL REVIEW D 88, 122003 (2013)

122003-10

J. Harms+ (2012)

/ 2224. 05. 2019 GWADW2019

Final(Target)

Phase-III (Now)

Development Plan

!4

Phase-I(2009)

10-8/√Hz @ 0.1 Hz(Established)

Principle Test

- Room Temp.- 25cm TM(s)

10-15/√Hz @ 0.1 Hz (Design)

Cryogenic Test

- Cryo. Temp. (4K)- 35cm TMs

10-19/√Hz @ 0.1 Hz(Target)

Goal

Phase-II(2015)

- Cryo. Temp. (4K)- 10m TMs

3.4 ねじれ振り子

図 3.3: 本実験で用いたねじれ振り子の写真

-26-

5 Experiments 5.1 Setup

Suspension Wires Mirrors forinterferometers

Magnets foractuators

Adjusting mass

Test Masses

Optical Bench

Figure 5.2: A picure of the detector.

59

potential sensitivity and scientific possibility in moderntechnologies.

Principle of a torsion-bar antenna.—A TOBA is com-prised of two bar-shaped test masses, arranged parallel tothe x-y plane and orthogonal to each other (Fig. 1). Eachbar is supported at its center, so as to rotate around thez axis. When GWs pass through this antenna, tidal forcesby the GWs will appear as differential angular changes inthese bars. These changes are extracted as a GW signal byusing a sensitive sensor, such as a laser interferometer.

The rotation angle ! of a test-mass bar from the originalposition is obtained by the equation of motion

I €!þ " _!þ #! ¼ FGWðtÞ; (1)

where FGW is the torque caused by the GWand " and # arethe damping factor and spring constant of the restoringtorque by the support, respectively. The moment of inertiaof the test mass is expressed as I ¼ R

$ðx2 þ y2ÞdV, where$ and V are the density and volume of the test mass,respectively. Assuming that the antenna is much smallerthan the wavelength of target GWs, the torque caused by aGW with an amplitude of hij is expressed as

FGWðtÞ ¼ 14q

ij €hijðtÞ; (2)

by using dynamic quadrupole moment q ij [16]. For barrotation, q 11 ¼ %q 22 ¼ %R

$ð2xyÞdV and q 12 ¼ q 21 ¼R$ðx2 % y2ÞdV.Here, we consider the response of a test-mass bar ar-

ranged along the x direction to GWs traveling along the zaxis, h11 ¼ %h22 ¼ hþ and h12 ¼ %h21 ¼ h&, where hþand h& denote the amplitudes of two independent polar-izations (plus and cross modes, respectively) of incidentGWs. In an approximation that the test-mass bar freelyrotate around the z axis (" ' 1 and # ' 1), Eq. (1) resultsin a simple equation, !1 ¼ %h&=2, where % is a shapefactor of the test mass; % ¼ q 12=I ’ 1 in the case of a thinbar. Another test-mass bar, arranged along the y axis,rotates with an opposite sign as !2 ¼ %%h&=2. The re-sultant output of the antenna is expressed as

!diffðtÞ ( !1 % !2 ¼ %h&ðtÞ: (3)

GWs with a cross polarization are observed as differentialangular fluctuations of the test-mass bars [17].

Now, we consider the situation that the antenna is rotat-ing around the z axis with an angular velocity of!rot. In an

approximation that rotation is sufficiently slow, the re-sponse of the antenna is expressed as

€! diff ¼ %½€h& cosð2!rottÞ þ €hþ sin ð2!rottÞ*; (4)

by calculating the torque, Eq. (2), in a coordinate rotatingwith the antenna. This indicates that the GW signal ismodulated by the rotation; a GW signal with an angularfrequency of !g is up- and down-converted to appear at!g + 2!rot frequencies. Equation (4) results in

!diff ’ %!!g

2!rot

"2½h& cosð2!rottÞ þ hþ sin ð2!rottÞ*; (5)

in the case of !g ' !rot. The low-frequency GW signal isup-converted to signals at an angular frequency of 2!rot.Equation (5) also shows that two polarization componentsof incident GWs are extracted from two quadrature phasesof the antenna output.Advantages of torsion-bar antenna.—A TOBA has sig-

nificant features in both ground-based and space-bornedesigns. As a ground-based antenna, a TOBA is a novelapproach to observe low-frequency GWs. In a usualground-based interferometric antenna, a test-mass mirroris suspended as a pendulum to behave as a free mass in thehorizontal plane. Conversely, it has almost no fundamentalsensitivity to GWs below the resonant frequency of thependulum (around 1 Hz). In a TOBA, a test-mass bar issupported as a torsion pendulum, with a low resonantfrequency on the order of a few millihertz in the rotationaldegree of freedom. Thus, a TOBA has a fundamentalsensitivity to low-frequency GWs.The modulation and up-conversion scheme by antenna

rotation is favorable for the observation of low-frequencyGWs below a few millihertz. Here, we note that the ob-servation run may be an intermittent one; the observationcan be a series of data-taking operations with rotation andreverse rotation. The up-conversion of the GW signal isalso advantageous from a practical perspective. Modu-lation prevents various types of low-frequency noises thatare difficult to suppress, such as drifts of instrumentscaused by daily or seasonal changes in the environmentand 1=f noises of electronics in sensors and controllers.As a space antenna, a TOBA has good compatibility

with spin-stabilized spacecraft. In a spacecraft, spinningitself is a simple and robust way to maintain its attitudewith a gyro effect, without additional disturbances fromattitude controllers. ATOBA, with its rotation axis alignedwith that of the spacecraft spin, has a wide observationband from the low-frequency limit determined by the ob-servation time. Another advantage in a space configurationis that the antenna is free from gravity-gradient and seismicnoises caused by ground motions.Sensitivity limits.—The fundamental sensitivity of a

TOBA is limited by the thermal fluctuation of the bars,readout noise of the angular motion, and effects of the barsupport, as detailed in Refs. [1,18]. We estimate the con-tributions of these noises in the case where a cylindrical baris suspended as a torsion pendulum at its center, and its

FIG. 1 (color online). Principle of a torsion-bar antenna. Twoorthogonal bars feel differential torques by incident gravitationalwaves.

PRL 105, 161101 (2010) P HY S I CA L R EV I EW LE T T E R Sweek ending

15 OCTOBER 2010

161101-2

M. OkadaMaster Thesis

A. ShodaPh.D Thesis

Test Masses

Optical Bench

Intermediate Masses

/ 29

Phase-III TOBA

!5

/ 2224. 05. 2019 GWADW2019

Overviews of Phase-III TOBA・Two-stage torsion pendulums・Active vibration isolation system・Cryogenic temperature (4K)

!6

1st Shield (50 K)

2nd Shield (4 K)

Cryogenic cooler

Test masses (35cm)

Intermediate masses

Optical Bench

/ 2224. 05. 2019 GWADW2019

Overviews of Phase-III TOBA

!7

1st Shield (50 K)

2nd Shield (4 K)

Cryogenic cooler

Active vibration isolation (AVIS)・ Isolation ratio ~ 102

@ 0.1 - 1 Hz・Reducing vibration caused

by the cooler via heatlinks

Cryogenic System・Cooled down to 4 K・Silicon/Sapphire wire

with High Q (Q ~ 108)

/ 2224. 05. 2019 GWADW2019

Readout SchemeCoupled-cavity wave front sensor (new idea)

・Compensate Gouy phase by auxiliary cavity

‣ HG10 mode resonates as well as HG00

‣ Induced HG10 is enhanced

‣ Higher sensitivity than normal WFS5×10-16 rad/√Hz @ 0.1 Hz

Optical configuration →

!8

Main Cavity Auxiliary CavityHG00

HG10

/ 2224. 05. 2019 GWADW2019

・Two shields:

‣ 1st shield: Aluminum, cooled to 50 K

‣ 2st shield: Cupper, cooled to 4 K・ Intermediate masses: cooled via heatlinks・Test masses: cooled via suspension wires

Cryogenic System

!9

1st shield

2nd shield

test massintermediate mass

4 K in 4 weeks

to AVIS

to cooler

1st Shield (50 K)

2nd Shield (4 K)

Test masses

Intermediate masses Heatlnks

/ 2224. 05. 2019 GWADW2019

Active Vibration Isolation・ Inertial sensor + Position sensor + Hexapod actuator (PZTs)・Function

‣ Reduce translational vibration @ the suspension point

‣ Reduce the vibration introduced by the cooler via the heatlinks

!10

10-7 m/√Hz@ 0.1 Hz

10-10 m/√Hz@ 1 Hz

Heatlnks

Hexapod (6 PZTs)

Inertial Sensor

Position Sensor

Requirement

Seismic Vibration (x, y, z)

Adiabatic Rod

/ 2224. 05. 2019 GWADW2019

Target Sensitivity of Phase-III TOBA

!11

RotationalSeismic

SuspensionThermal

CoatingThermal

ShotNoise

TranslationalSeismic

Target Sensitivity

/ 2224. 05. 2019 GWADW2019

Science of Phase-III TOBA

1 Mpc for 105 M⦿

!12

IMBH BinaryMerger

GW StochasticBackground

NewtonianNoise

EarthquakeAlert

Astrophysics

Phase-III TOBA

Geophysics

100 km for M6.0 in 10 sec

/ 29

Current Status

!13

/ 2224. 05. 2019 GWADW2019

Current Status・Cryogenic System

◦ 2 shields were installed◦ 1st cryogenic torsion pendulum test◦ Some problems about cooling

・AVIS◦ Construction were done◦ Seismic vibration isolation were tested◦ Tilt-horizontal coupling limits the performance

!14

/ 2224. 05. 2019 GWADW2019

・Shields and suspension system were installed (2019 Feb.)・Suspension is a temporary setup to test cooling

‣ Suspension wire: CuBe

‣ Heatlinks are connected to both IM and TM

Cryogenic System

!15

IM

1st shield

2nd shield

TM

IM

OpticalLever

/ 2224. 05. 2019 GWADW2019

Result of Cooling1st cryogenic torsion pendulum test・Cooled down to 8.5K

‣ There exists some kind of heat injection

・ It took ~ 2weks to reach equilibrium

‣ Cooling speed is twice slower as expected

Further investigation is needed

!16

1st shield

2nd shield

TM

IM

Measured in 2019 Apr.

/ 2224. 05. 2019 GWADW2019

Sensitivity

!17

ADC Noise

Mechanical Mode(Pitch, Roll, Pendulum, …)

Mechanical Mode(Yaw)

UnknownNoise

Measured in 2019 Apr.

/ 2224. 05. 2019 GWADW2019

AVIS Performance・Constructed separately from the cryogenic system・Succeeded in reducing Seismic motion by 1/100 @ 1Hz・Current problem: tilt-horizontal coupling of inertial sensor

!18

Inertial Sensor

Hexapod

Seismic motion(Free)

Seismic motion (Controlled)

Requirement

1/100

Measured in 2019 Apr.

x

z

y

/ 2224. 05. 2019 GWADW2019

AVI Performance

・Controlled in each DoF independently

!19

1/100

1/3001/300

/ 2224. 05. 2019 GWADW2019

Tilt-Horizontal Coupling・When controlling one translational direction, we also shake in

Pitch/Roll rotation・Below 0.1 Hz tilt signal exceeds true translational signal

‣ Cannot put blend frequency below than 0.1 Hz

‣ Cannot control at the same time

!20

y (when x is controlled)

shaking in Roll

ideal

ideal

Shaking in Pitch

Responcefrom actuators

to inertial sensor(x)

/ 2224. 05. 2019 GWADW2019

Future WorksAVIS・Combine with cryogenic system・Test the reduction of the vibration introduce from heatlinks・ Introduce tilt meter to reduce tilt-horizontal coupling by controlling

tilt motion

Cryogenic system・Check why the cooling speed is slower than expected・Develop high Q wire at 4K

Others・Construct optical system・Measure how the cooling affects components (how much they drift)

!21

/ 2224. 05. 2019 GWADW2019

Summary・Phase-III TOBA is under development

‣ Target sensitivity: 10-15/√Hz @ 0.1 Hz

‣ Science target: Newtonian noise, earthquake alert, IMBH

・Main Issue:

‣ Active vibration isolation system

‣ Cryogenic systemboth system are constructed and evaluated

・Future works

‣ Develop high Q value wire

‣ Improve cooling speed and reduce heat injection

‣ Introduce tiltmeter to avoid tilt-horizontal coupling

‣ Combine AVIS and cryogenic system together!22