Toward the Advanced LIGO optical configuration investigated in 40meter prototype

Transcription

1 Toward the Advanced LIGO optical configuration investigated in 4meter prototype Aspen winter conference Jan. 19, 25 O. Miyakawa, Caltech and the 4m collaboration LIGO- G547--R Aspen winter conference, January 25 1

2 Caltech 4 meter prototype interferometer Objectives Develop lock acquisition procedure of detuned Resonant Sideband Extraction (RSE) interferometer, as close as possible to Advanced LIGO optical design Characterize noise mechanisms Verify optical spring and optical resonance effects Develop DC readout scheme Extrapolate to AdLIGO via simulation etc. LIGO- G547--R Aspen winter conference, January 25 2

3 Advanced LIGO optical configuration LIGO:Power recycled FPMI» Optical noise is limited by Standard Quantum Limit (SQL) AdvLIGO:GW signal enhancement using Detuned RSE» Two dips by optical spring, optical resonance» Can overcome the SQL QND detector Standard Quantum Limit FP cavity Laser PRM Power FP cavity Detuning BS GW signal LIGO- G547--R Aspen winter conference, January 25 3

4 Differences between AdvLIGO and 4m prototype 1 times shorter cavity length Arm cavity finesse at 4m chosen to be = to AdvLIGO» Storage time is x1 shorter. Control RF sidebands are 33/166 MHz instead of 9/18 MHz» Due to shorter PRC length. LIGO-I 1-watt laser, negligible thermal effects» 18W laser will be used in AdvLIGO. Noisier seismic environment in town» >1x1-6 m at 1Hz Smaller stack, commercial active seismic isolation» STACIS isolators in use on all test chambers, providing ~3 db of isolation from 1-1 Hz. LIGO-I single pendulum suspensions are used» AdvLIGO will use triple (MC, BS, PRM, SRM) and quad (ITMs, ETMs) suspensions. LIGO- G547--R Aspen winter conference, January 25 4

5 Pre-Stabilized Laser(PSL) and 13m Mode Cleaner(MC) AP BS ITMx X arm ITMy MOPA126 FSS PSL SP 13m Mode Cleaner PO Y arm LIGO- G547--R PMC AOM VCO 13m MC 4m arm cavity 1W MOPA126 Frequency Stabilization Servo (FSS) Pre-Mode Cleaner (PMC) 13m Mode Cleaner with digital controlled suspension Good noise performance and stable operation Aspen winter conference, January 25 5

6 In-vacuum Faraday Isolator and In-vacuum Mode Matching Telescope Faraday Isolator IFO MC3 MMT FI MC1 Mode Matching Telescope LIGO- G547--R Aspen winter conference, January 25 6

7 LIGO-I type single suspension Each optic has five OSEMs (magnet and coil assemblies), four on the back, one on the side The magnet occludes light from the LED, giving position Current through the coil creates a magnetic field, allowing mirror control LIGO- G547--R Aspen winter conference, January 25 7

8 STACIS Active seismic isolation One set of 3 for each of 4 test chambers 6-dof stiff PZT stack Active bandwidth of.3-1hz, 2-3dB of isolation passive isolation above 15 Hz. LIGO- G547--R Aspen winter conference, January 25 8

9 LIGO- G547--R Digital control system Aspen winter conference, January 25 9 Demodulated signal from PD Feedback filters Output to suspensions

10 Signal extraction for AdvLIGO f 1 f 2 PRM 4km ETMy ITMy BS ITMx 4km ETMx Two modulations are used to separate high finesse, 4km long arm cavity signals from Central part (Michelson, PR, SR) signals. Only + f 2 is resonant on SRC Unbalanced sidebands of +/-f 2 make error signal of Central part SRM Carrier (Resonant on arms) Single demodulation Arm information -f 2 -f 1 f 1 f 2 Double demodulation Central part information Arm cavity signals are extracted from beat between carrier and f 1 or f 2. Central part (Michelson, PR, SR) signals are extracted from beat between f 1 and f 2, not including arm cavity information. LIGO- G547--R Aspen winter conference, January 25 1

11 5 DOF for length control Signal Extraction Matrix (in-lock) ETMy Port SP Dem. Freq. f 1 L + 1 L -3.8E-9 l E-3 l -1.3E-6 l s -2.3E-6 AP f 2-4.8E E-8 1.3E-3-1.7E-8 L y SP f 1 f 2-1.7E-3-3.E E-2-1.E-1 AP f 1 f 2-6.2E-4 1.5E-3 7.5E E-2 PO f 1 f 2 3.6E-3 2.7E-3 4.6E-1-2.3E-2 1 ITMy Laser PRM l y l sy BS ITMx ETMx SP l x PO l sx SRM AP L x Common of arms : L + =( L x + L y ) / 2 Differential of arms : L = L x L y Power recycling cavity : l + =( l x + l y ) / 2 Michelson : l = l x l y Signal recycling cavity : l s =( l sx + l sy ) / 2 LIGO- G547--R Aspen winter conference, January 25 11

12 Disturbance by sidebands of sidebands Original concept Real world Carrier Carrier -f 2 -f 1 f 1 f 2 -f 2 -f 1 f 1 f 2 Sidebands of sidebands are produced by two series EOMs. Beats between carrier and f 2 +/- f 1 disturb central part. Port Dem. Freq. L + L l + l l s SP f E-8-1.2E-3-1.3E-6-6.2E-6 AP f 2 1.2E E-5 1.3E-3 6.5E-6 SP f 1 f E E-2-1.1E-1 AP f 1 f 2-5.7E E E-2 PO f 1 f E-1-3.5E-2 1 LIGO- G547--R Aspen winter conference, January 25 12

13 Mach-Zehnder interferometer on 4m PSL to eliminate sidebands of sidebands Series EOMs with sidebands of sidebands f 1 f 2 EOM1 EOM2 Mach-Zehnder interferometer no sidebands of sidebands from beginning PMC trans f 2 EOM2 f 1 PZT Locked by internal modulation To MC PMC transmitted EOM1 PD PD BS1 166MHz EOM 33MHz EOM PZT mirror BS2 29MHz EOM to MC LIGO- G547--R Aspen winter conference, January 25 13

14 MZ eliminates sidebands on sidebands MCT light, series EOMs carrier 33 MHz 166 MHz 199 MHz SBonSB carrier parallel EOMs in MZ ifo No sidebands on sidebands! (hard to directly compare because we can t turn the modulation depth up as high as we could before; but we can get up to Γ =.25 easily) LIGO- G547--R Aspen winter conference, January 25 14

15 Important Milestones September, 23 Four TMs and BS: installed November 23 FP Michelson locked February 24 Power Recycling Mirror (PRM), Signal Extraction Mirror (SRM) installed June 24 Mach-Zehnder installed August 24 DRMI locked with carrier resonance October 24 DRMI locked with sideband resonance November 24 Off-resonant lock of arm cavities with DRMI Bright port PRM Y arm BS SRM Dark port X arm LIGO- G547--R Aspen winter conference, January 25 15

16 LIGO- G547--R Aspen winter conference, January 25 16

17 Lock Acquisition of Detuned RSE 1. lock central part 2. lock arm cavities ITMy ETMy PRM BS ITMx ITMy Step 2 Step 1 Step 3 SRM PRM BS SRM ITMx ETMx Central part: not disturbed by carrier resonance on arm cavity (but disturbed by sidebands resonance) Lock acquisition After lock: l - : 12 Hz DDM@AP l + : 33MHz@SP DDM@SP l s : DDM@PO DDM@PO LIGO- G547--R Arm cavities: not disturbed by locked central part Lock each arm cavity independently Switch control servo to common/differential control Aspen winter conference, January 25 17

18 l - signal with double demodulation Lock point Lock point Good l - signal when l + and l s is locked No good l - signal once l + and l s start moving LIGO- G547--R Aspen winter conference, January 25 18

19 l - signal with double demodulation Lock point Lock point Good l - signal when l + and l s is locked No good l - signal once l + and l s start moving LIGO- G547--R Aspen winter conference, January 25 19

20 Looking for good signal for lock acquisition Unfortunately, no way to lock central part directly using the original double demodulation Dither locking for l - signal Divide signal by inside power» Good cancellation of power recycling V l = d d l V = AP V V V -V AP PO PO AP 2 VPO V PO Laser SP PRM PO ITMy BS SRM AP 1 ITMx V l ~ LPF LPF LPF LPF a few khz V PO (V PO ) V AP (V AP ) Digital calculation LIGO- G547--R Aspen winter conference, January 25 2

21 l - signal with dither Lock point Dither on ITMx, ITMy with 1kHz Lock point Error signal is calculated digitally as follows; d VAP V AP VPO -VAP V PO V l 2 d l V = = PO VPO l - signal does not depend on l + at all l - dither locking signal gain depends on l s, but polarity of signal is always the same LIGO- G547--R Aspen winter conference, January 25 21

22 With l- dither-locked, there s always a good l+ signal, for all values of ls. The locking point may not be at l+ = º! The PRM follows the swinging of the SRM; this signal keeps the combined cavity locked. Then, once ls is locked, we ll recover l+ = º. Lock l+ with DDM at SP LIGO- G547--R Aspen winter conference, January 25 22

23 l s signal with l - and l + lock Good l s signal can be extracted once l + is locked to zero-crossing point LIGO- G547--R Aspen winter conference, January 25 23

24 DRMI lock with Unbalanced sideband by detuned cavity August 19, 24 DRMI locked with carrier resonance (like GEO configuration) November 9, 24 DRMI locked with sideband resonance (Carrier is anti resonant preparing for RSE.) November 16, 24 Switched to DDM control Carrier Can be locked with DDM directly Longest lock: 2.5 hours Typical lock acquisition time ~1sec Carrier 33MHz 166MHz ITMy BS ITMx Unbalanced 166MHz 33MHz PRM DDM PD DDM PD SRM OSA DDM PD LIGO- G547--R Belongs to next carrier Belongs to next carrier Aspen winter conference, January 25 24

25 Trial of Arm lock with DRMI 1. Disturbance of sideband resonance on arm cavities» Gain and amplitude limitter on DRMI control 2. Slow digital sampling rate 3. Coupling between X-arm and Y-arm signal through carrier resonance on Michelson part» Off-resonant lock for carrier ETMy PRM ETMy ITMy BS Off-resonant arm lock for carrier ITMx ETMx Carrier 33MHz 166MHz ETMy Sutter PRM Sideband resonance ITMy BS ITMx ETMx SRM ETMy DRMI Sutter PRM ITMy BS ITMx ETMx SRM Sutter PRM ITMy BS Lock with Carrier resonance ITMx ETMx Sutter SRM LIGO- G547--R Aspen winter conference, January 25 SRM 25

26 Off-resonant lock scheme for arm cavity Transmitted light is used as 1 + offset Transmitted power Off-resonant Lock point Resonant Lock to avoid coupling of carrier in Michelson part when arm cavity is locked. LIGO- G547--R Aspen winter conference, January 25 26

27 Off resonant Arm lock with DRMI November 25, 24 Both arms locked with DRMI Off-resonant carrier on arm cavities Last < 1 min Locked only 2 times DRMI with single arm lock Not so difficult Last ~1 min Lock acquisition time ~1 min Reducing offset starts oscillation caused by optical lever servo, under investigation Arm power Error signal Ideal lock point Yarm lock Offset lock Xarm lock Offset lock LIGO- G547--R Aspen winter conference, January 25 27

28 Summary Optical configuration for AdvLIGO being developed at 4m prototype interferometer Stable operation of PSL and MC Locking of FPMI with digital LSC system (misaligned PRM, SRM), measurement of displacement noise Sidebands of sidebands: eliminated by M-Z interferometer Guided locking of DRMI using Dither-locking with carrier/sideband resonance Locking of DRMI with DDM with sideband resonance Off-resonant locking of both arms with DRMI (not perfect but very close to final configuration) Hope we succeed in locking full RSE very soon! LIGO- G547--R Aspen winter conference, January 25 28

29 LIGO- G547--R Aspen winter conference, January 25 29

30 4m vs. Ad-LIGO 4m x6 x1.5 x3 Ad-LIGO x2 x6 x17 LIGO- G547--R Aspen winter conference, January 25 3

31 Abs +33 off-resonance +33 resonance +166resonance Design Lock point SP33 Lock point.56 degree LIGO- G547--R Aspen winter conference, January 25 31

32 Original design (no offset) resonance 25 Double Demodulation at SP Abs -.4 dc= l+ l- ls : off-resonant -33 : off-resonant +166: resonant -166 : anti-resonant resonance SP33 Lock point.56degree Design Lock point 1 Demodulation Phase of f2 l + and l s plot separated Difficult Aspen winter conference, January Demodulation Phase of f1 LIGO- G547--R

33 .2 l degree Double Demodulation at SP resonance resonance Abs dc= l+ l- ls : resonant -33 : resonant +166: off-resonant -166 : anti-resonant SP33 Lock.2 point.56degree Design Lock point Demodulation Phase of f2 l + and l s plot overlapping DC line changed Difficult Aspen winter conference, January Demodulation Phase of f1 LIGO- G547--R

34 Double demodulation signal of l + What we expected» Big offset when cavity is not locked» No disturbance of carrier What we have seen» No offset» Big disturbance of carrier Carrier resonance No offset LIGO- G547--R Aspen winter conference, January 25 34

35 Double Demodulation Double Demodulation at SP -.4 dc= l+ l- ls Demodulation Phase of f Double Demodulation used for l +, l -, and l s Demodulation phases optimized to suppress DC and to maximize desired signal [S.Kawamura, Signal Extraction Matrix of the 4m Detuned RSE Prototype, LIGO-T41--R (24)] Demodulation Phase of f1 Aspen winter conference, January LIGO- G547--R

36 LIGO- G547--R Gain of dither locking signal l- dither locking signal gain depends strongly on ls But polarity of signal is always the same Can handle this with a limiter l- dither locking signal doesn t depend on l+ at all! Signal is degraded by presence of RF sidebands turn them down low (Γ<.2) to acquire dither lock, then ramp them back up. error signal gain of l- error signal l s Aspen winter conference, January l -

37 Optical configuration for Gravitational wave interferometer Gravitational wave detection using Michelson interferometer Signal and power enhancement using Fabry-Perot cavity in each arm FP cavity Laser FP cavity BS Power enhancement using Power Recycling FP cavity Laser PRM FP cavity BS LIGO- G547--R Aspen winter conference, January 25 37

38 Once we acquire full lock Measure in-lock transfer functions. Verify optical resonance and optical spring Begin noise characterization Operate at a different SEC tune? Begin work towards DC readout» output mode cleaner» offset locking of arms» Tuning of homodyne phase LIGO- G547--R Aspen winter conference, January 25 38

39 Contents Over view of 4meter prototype Signal extraction for Advanced LIGO Lock acquisition of Dual Recycled Michelson (DRMI) Off-resonant lock of arm cavities with DRMI LIGO- G547--R Aspen winter conference, January 25 39

40 LIGO- G547--R Aspen winter conference, January 25 4

41 Dual Recycling Summation cavity for End to End model Calculation time is determined by shortest cavity (Michelson) length. Calculating many time steps at once in Michelson part Summation cavity E( t) = M E( t τ ) = M Μ = M 2 N LIGO- G547--R E( t 2τ ) τ = E( t Nτ ) L N = l arm 4 times faster for LIGO 4 times faster for 4meter MI L c arm >> N step 1 step l c Mi M = M + δm M : matrix for DR summation cavity M δm δm M M : scalar for PR summation cavity M δm = δm M Aspen winter conference, January 25 41

42 4m Team On the payroll: Ben Abbott, Osamu Miyakawa, Bob Taylor, Steve Vass, Alan Weinstein Grad students: Lisa Goggin, Rob Ward LIGO engineering support: Jay Heefner, Rolf Bork, Alex Ivanov, Flavio Nocera, Michael Smith, Lisa Bogue, many others Visitors: Seiji Kawamura, Fumiko Kawazoe, Shihori Sakata, Bryan Barr, Sascha Schediwy, Kentaro Somiya, Rana Adhikari ( Hartmut Grote of GEO arrives Oct 1) LIGO- G547--R Aspen winter conference, January 25 42

43 LIGO- G547--R Gain of dither locking signal l- dither locking signal gain depends strongly on ls But polarity of signal is always the same Can handle this with a limiter l- dither locking signal doesn t depend on l+ at all! Signal is degraded by presence of RF sidebands turn them down low (Γ<.2) to acquire dither lock, then ramp them back up. error signal gain of l- error signal l s Aspen winter conference, January l -

44 First lock of Dual recycled Michelson (demod phase for l s ) l s lock DC@SP l + lock Error of Dither 33MHz@SP DC@AP l - lock DDM@SP (demod phase for l + ) Lock now: Control later: l - : DDM@AP l + : 33@SP DDM@SP l s : DDM@SP DDM@PO l s lock at -5.2 sec l - lock at -5.3 sec l + lock at -5.6 sec LIGO- G547--R Aspen winter conference, January 25 44

45 Control room LIGO- G547--R Aspen winter conference, January 25 45