Fumiko Kawazoe, Shuichi Sato, Volker Leonhardt, Osamu Miyakawa

LSC meeting, Hannover,10/24/2007 1

Fumiko Kawazoe, Shuichi Sato, Volker Leonhardt, Osamu Miyakawa

Tomoko Morioka, Atsushi Nishizawa, Seiji Kawamura, Akio Sugamoto

Ochanomizu University NAOJ/Caltech Tokyo University Kyoto University

Experimental Demonstration of a control Experimental Demonstration of a control scheme for a tuned RSE interferometer for scheme for a tuned RSE interferometer for

next-generation gravitational-wave next-generation gravitational-wave detectorsdetectors


Motivation

• Plans to use an RSE interferometer in some of the next-

generation interferometers for better sensitivities.– Adv.LIGO…Detuned RSE

– LCGT…Tuned RSE ( = zero detuning or broadband )

• Controlling DOFs is vital for a detector– Control scheme developed for LCGT

– Can be a back-up design for Adv.LIGO(detuned)

• Prototype experiment to verify the control scheme


Controlling the RSE


Ｌｙ

Ｌｘlaser

•L+ = (Lx + Ly) / 2

•L- = (Lx - Ly) / 2 Ｌｙ

Ｌｘ

•l+ = (lx + ly) / 2

•l- = (lx - ly) / 2

ｌｘ

ｌｙ

•ls = (lsx + lsy) / 2

ｌｓｘ

ｌｓｙ

Fabry-Perot cavity

BS

PRM

SEM

Fabry-Perot cavity

5 degrees of freedoms

A very complicated control system due to the increased number of DOFs


Control scheme concepts•Fabry-Perot cavities’ control signal---beat between the carrier and PM sidebands•Central part of the RSE--- beat between the AM and PM sidebands

•Fabry-Perot control signal are bigger by the high finesse

•Can separate FP control signal and the central control signal by not using the carrier for the central part


The central part control strategy

Due to the Michelson AsymmetryAM all reflect from the Michelson partPM all transmit though the Michelson part

AM resonant inside PRCPM resonant inside PRC+SEC

Contrasting behavior in the Michelson part as little as possible signal coupling

For AM sidebands: 2δl = nλ(n = 1,2,3…)

For PM sidebands: 2δl = (2m+1)/2 λ (m= 0,1,2…)

--- Michelson Asymmetry

Cavity length design


The sensing signal matrix

11 8.0 e-6 -2.6 e-2 6.2 e-4 1.3 e-2 8.0 e-6 -2.6 e-2 6.2 e-4 1.3 e-2

-1.5 e-1 -1.2 e-2 1.1 -2.2 e-2 1

-1.0 e-4 7.6 e-2 1.4 e-3 1 1.1 e-5

-4.9 e-2 -1.1 e-4 1 -8.6 e-3 -5.3 e-1

-2.2e-8 1 1.4 e-8 1.3 e-2 2.0 e-8

Fabry-Perot signals are clear from other signals.

Linearly independent matrix.

Theoretical signal matrix (DC)With the prototype parameters

L+ L- l+ l- ls

BP

DP

BP(DD)

DP(DD)

PO(DD)

These 6 are small


Prototype RSE experiment

3rd chamber

Built inside NAOJ’s campus in 2005

RSE locked in 2007

ETMiITMp

mm lens


PM ：17.25MHzAM ：103.5MHz

~90mW

∞ 6.0m

Stable cavities5.7m

FP Finesse ：～ 100

（ specialized for lock

demonstration ） 4.34m

Optical layout, and Parameters


Mainly Input table, plus some detection ports

LASER

EOM(AM)

EOM(PM)¼ λ

PBS

TRp PD

MZ PD

l+ PD

L- PDOSA


Detection ports

To OSA

l- PD

L- PD

Tri PD

DP monitor

¼ λPBS

ls PD


Damping magnets

Copper

Upper mass

Eddy current damping system

Double suspension system

Two loop

Single loop

30cm

Qpitch:3

Qyaw:7

Suspension system


Test mass

Actuator magnets x4

Actuator coils x4

Current in

Suspension systemDC alignment


LASER

BP

DP

State1: l- lockedState2: l- & l+ lockedState3: l-, l+, & ls lockedState4:RSE locked

　 AM　 PM

Lock of the tuned RSE: Lock state

State0: Nothing is locked

s

LL

LL


-40 -30 -20 -10 0 10 20 30 400

5

10

15

time[s]

DC

at

DP

- 40 -30 -20 -10 0 10 20 30 40

1234

time[s]

DC

at

BP

- 40 -30 -20 -10 0 10 20 30 400

5

10

time[s]

PR

C g

ain

- 40 -30 -20 -10 0 10 20 30 4002468

time[s]

DC

at

Tri

- 40 -30 -20 -10 0 10 20 30 4002468

time[s]

DC

at

Trp

0: not controlled 1: l- locked 2 : l-, l+ locked 3: l-, l+, ls (central part) locked 　 4:RSE locked (longest ~15min. disturbed by human activity)

0 0&1 2 3 4

10sec

DP

BP

PR

Cgai

n

FP

iF

Pp

Pow

er

（AU

）

Pow

er

（AU

）

Time(sec)

state

Lock of the tuned RSEDC power at various detection portsl- locked

l+ locked ls locked L+ & L- locked

16

-100 -50 0 50 1000

0.1

0.2

0.3

0.4

0.5

freqnency[MHz]

Pow

er[

arb.]

-100 -50 0 50 1000

0.1

0.2

0.3

0.4

0.5

freqnency[MHz]-100 -50 0 50 100

0

0.1

0.2

0.3

0.4

0.5

freqnency[MHz]

-100 -50 0 50 1000

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

freqnency[MHz]

Pow

er[

arb.]

-100 -50 0 50 1000

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

freqnency[MHz]-100 -50 0 50 100

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

freqnency[MHz]

　 PM

　 AM

PO

DP

PM

Optical Spectrum Analyzed at PO, and DP

AM

0&1 2 3

( 1: l- 　 2: l-, l+ 　 3: central part )

Lock of the tuned RSE:Evaluation of lock


L+ L- l+ l- ls

1 5.6 e-2 -5.0 e-2 3.5 e-2 1.1 e-21 5.6 e-2 -5.0 e-2 3.5 e-2 1.1 e-2

-6.5 e-1 3.3 e-1 9.2 e-1 -3.6 1

4.6 e-2 -1.6 e-1 1.6 e-1 1 6.2 e-2

-2.8 e-1 -7.1 e-2 1 -3.6 e-1 -1.5 e-1

4.5e-2 1 1.3 e-2 -8.3 e-2 9.7 e-3

BP

DP

BP(DD)

DP(DD)

PO(DD)

Measured signal matrix （ preliminary ）

•Dem. phase set to maximize the desired signal

•Source:3.3kHz

L+ L- l+ l- ls

Needs better lock quality for further discussion..

18

Control of a tuned RSE interferometer was successfully demonstrated.　　　

Results and future work

•FP cavities with L+, and L-.

•Central part with double mod-demod.

Lock acquisition scheme for LCGT is verified with this control scheme.

Future work:•Further investigation of the signal matrix (further diagnolazation, possible application of the delocation scheme, shot noise sensitivity as a detector after the lock is acquired)•Measurement of the optical gain matrix with better lock quality.


Delocation scheme


Diagnolized signal matrix

Documents

Fumiko Kawazoe, Shuichi Sato, Volker Leonhardt, Osamu Miyakawa