1. Present Status of the Virgo Experiment
Raffaele Flaminio
EGO - CNRS/IN2P3
Summary
I. The Virgo experiment
- Detector design
- The actors
- Main construction steps
II. Detector commissioning
- Commissioning plan overview
- Main results
Next steps
III. Data analysis preparation
- Conclusions
Cascina, July 8th, 2004

2. I. The Virgo experiment
Cascina, July 8th, 2004

3. Virgo: Interferometer design
Cascina, July 8th, 2004

4. Virgo: Laser 20W, Nd:YVO4 laser, two pumping diodes
Injection locked to a 1W Nd:YAG master laser
Required power stability: 10-8
Required frequency stability: mHz
Cascina, July 8th, 2004

5. Virgo: Input Mode-Cleaner
Input mode-cleaner: curved mirror
Input mode-cleaner: dihedron
Cascina, July 8th, 2004

6. Virgo: Mirrors Material: fused silica
Dimension: 35 cm diameter, 10 cm thick
Mass: ~21 Kg
Surface deformation < l/100
Anti-reflective coating ~ 10-4
Cascina, July 8th, 2004

7. Virgo: Suspensions Multi stage pendulum
pre-isolator: inverted pendulum
cantilever springs for vertical isolation
cascade of six stages
height ~ 10 m
Expected attenuation > 10 Hz
Cascina, July 8th, 2004

8. Virgo: Output Optics Output Mode-Cleaner Detection bench
Light filtering with output mode-cleaner
3.6 cm long monolithic cavity
Light detection with InGaAs photodiodes
3 mm diameter, quantum efficiency > 90%
Detection bench
Cascina, July 8th, 2004
Output Mode-Cleaner

9. Virgo: Planned Sensitivity
First attempt to extend detection band down to 10 Hz !
Cascina, July 8th, 2004

10. The Virgo collaboration
French-Italian collaboration supported by CNRS in France and INFN in Italy
Composed by 11 laboratories:
About 170 people participated to the construction of Virgo
Main responsibilities: INFN Pisa Seismic isolation, Vacuum
INFN Florence/Urbino Seismic isolation
INFN Frascati Alignment
INFN Napoli Environmental monitoring, Alignment
INFN Perugia Mirror suspension
INFN Rome Mirror suspension and Control
LAPP Annecy DAQ, Detection, Vacuum chambers LAL Orsay Global control, Tube, Baffles OCA Nice Laser and Injection IPNL Lyon Mirror coatings
ESPCI Paris Mirror metrology
LAPP Annecy, INFN Firenze, INFN Frascati, IPN Lyon,
INFN Napoli, Observatoire de Nice, LAL Orsay, ESPCI Paris,
INFN Perugia, INFN Pisa, INFN Roma
Cascina, July 8th, 2004

11. The construction of Virgo
Planning driven by the construction of the infrastructures
Construction of central area
Construction of the arms and the terminal buildings
Installation and commissioning of the central interferometer
Vacuum tubes installation
Installation of final mirrors and of terminal suspensions
….. Detector commissioning
‘Central area’
available since 1998
Cascina, July 8th, 2004

12. European Gravitational Observatory
What is EGO ?
Consortium settled up by the CNRS and the INFN
Main objectives of EGO ?
1) to support the commissioning of Virgo, its operation, maintenance and upgrades
2) to create and run a computing center for the analysis of data,
3) to ensure the maintenance of the site and the related infrastructures
4) to promote R&D useful for the detection of gravitational waves
Large support given to the commissioning activities:
Support to most of the main Virgo sub-systems
Leading role in: alignment, injection control, suspension control, data storage and vacuum
Interferometer operation
Cascina, July 8th, 2004

13. II. Detector commissioning
Cascina, July 8th, 2004

14. Commissioning plan overview
Phase A: Commissioning of interferometer arms
Phase B: Commissioning of the recombined interferometer
Phase C: Commissioning of the recycled interferometer
Status today:
- phase B near to completion
- moving to phase C
Cascina, July 8th, 2004

15. Commissioning plan overview
Phase A: Commissioning of interferometer arms
Phase B: Commissioning of the recombined interferometer
Phase C: Commissioning of the recycled interferometer
Status today:
- phase B near to completion
- moving to phase C
Cascina, July 8th, 2004

16. Commissioning plan overview
Phase A: Commissioning of interferometer arms
Phase B: Commissioning of the recombined interferometer
Phase C: Commissioning of the recycled interferometer
Status today:
- phase B near to completion
- moving to phase C
Cascina, July 8th, 2004

17. Commissioning plan overview
Phase A: Commissioning of interferometer arms
Phase B: Commissioning of the recombined interferometer
Phase C: Commissioning of the recycled interferometer
Status today:
- phase B near to completion
- moving to phase C
Cascina, July 8th, 2004

18. Cavities alignment & locking
Power fluctuations due to:
- laser frequency noise (above few Hz)
- mirror angular motions (below few Hz)
As soon as injection and north arm suspensions became available (September 2003)
1) Cavity alignment
2) North cavity locking using B1’
3) Output mode-cleaner locking
4) North cavity lock using B1
Cascina, July 8th, 2004

19. Laser frequency noise reduction
Laser frequency noise induced by input mode-cleaner control noise
Noise reduced when control gain/bandwidth reduced
Cascina, July 8th, 2004
Peaks height might be a problem for recycling locking ..… work in progress

20. Automatic alignment: principle
Anderson technique:
- Modulation frequency coincident with cavity TEM01 mode
- Two quadrants looking at the cavity transmission (at two different Guoy phases)
- Four signals to control the 2x2 mirror angular positions (NI & NE)
Cascina, July 8th, 2004

21. Automatic alignment: results
Automatic alignment operated on both arms
- bandwidth ~ 3-4 Hz
- control precision ~ 0.1 mrad rms
It allows to:
- switch completely OFF local controls on all four cavity mirrors
stabilize power stored in the cavity
increase locking duration
Cascina, July 8th, 2004

22. ‘Tide control’ Earth tide > Mirror actuators range ( ~ 100 mm)
Move low frequency component of the correction to the top of the inverted pendulum
Cavity transmission
Correction to the mirror
Suspension point position
6 h
Cavity transmission
Correction to the mirror
Suspension point position
24 h
Cascina, July 8th, 2004

23. Laser frequency stabilization
First test with North arm cavity
- North cavity error signal sent to the input mode-cleaner (below 200 Hz) and to the laser (above 200 Hz)
- Reference cavity error signal used to control cavity length at DC
Cascina, July 8th, 2004

24. Commissioning runs: C1, C2 & C3
Three commissioning runs performed since November ( ~ 1 each 2 months)
Each run last 3-4 days
Goals:
- Verify ITF cavities performances on longer time scales
- Provide real data to the collaboration
C1 (14-17/11/2003) - North cavity and OMC locked
C2 (20-23/02/2004)
- C1 + Automatic alignment
- West arm locked
C3 (23-27/04/2004)
- C2 + Laser frequency stabilization C1 C2 C3
Cascina, July 8th, 2004

25. Interferometer lock acquisition
Each cavity locked independently
Dark fringe locked with B2 or B1’
Main outcome:
- B1’ signal very sensitive to alignment
- due to the use of Anderson technique
- effect disappear if B1 is used
- simulation work in progress to understand what will happen in the recycled interferometer
First lock on Feb the 27th
Cascina, July 8th, 2004

26. Status of recombined interferometer today
Interferometer locked with B1 and B2
Output mode-cleaner locked to the TEM00 mode
Automatic alignment on both arms
Tidal control on both arms
Laser frequency stabilized to cavities common mode
ITF common mode locked to reference cavity
Cascina, July 8th, 2004

27. C4 run: Summary
Configuration: recombined ITF with ‘nearly’ complete control system
Duration: 5 days, June 2004 (last week)
Test periods at the beginning and at the end of the run
9 losses of lock during quiet periods (all understood, one due to an earthquake in Alaska)
Longest locked period: ~ 28 h, relatively stable noise level
Cascina, July 8th, 2004

28. C4 run: ITF sensitivity ‘without power recycling’ C3 (April 2004)
C4 (June 2004)
Cascina, July 8th, 2004

29. C4 run: Noise Sources
Cascina, July 8th, 2004

30. Next steps Lock recycled interferometer
Planned strategy: based on the LIGO experience WE NE
Present status:
- State 2: locked
- State 3: locked for 10 min with WE mirror misaligned
Potential problem due to light backscattered inside the input mode-cleaner
Short term solution: reduce light entering the interferometer
Mid-term solution: add optical isolation between the interferometer and the input mode-cleaner
Next week: try to lock state 4
Cascina, July 8th, 2004

31. III. Data analysis preparation
Cascina, July 8th, 2004

32. Data analysis organization
Six working groups settled up inside Virgo since 1998
h Reconstruction: chair F.Marion (LAPP Annecy)
Noise analysis (data quality): chair J.Y.Vinet (OCA, Nice)
Coalescing binaries: chair A.Vicere (INFN Florence/Urbino)
Burst: chair P.Hello (LAL, Orsay)
Periodic sources: chair S.Frasca (INFN Rome)
Stochastic background: chair G.Cella (INFN Pisa)
Hardware and Software infrastructure developed for On-line/In-Time analysis:
h Reconstruction
Data quality
Coalescing binaries
Burst
Cascina, July 8th, 2004

33. On-line/In-time analysis
Software architecture
All parts available
Each box runs properly in standalone mode
Connection between all boxes running
Full connection has to be tested under realistic conditions
Connection to DAQ foreseen in the next fall
Hardware architecture
300 GFlops PC’s farm being deployed at EGO
Mainly needed for coalescing binaries search
Will be fully available next year
Line Removal
Whitening
Trigger Manager
Merlino Burst
Merlino CB
MBTA
h reconstruction
Cascina, July 8th, 2004

34. Mock Data Challenge Main purpose: test the full In-Time chain
3 MDC since the beginning of 2004 (January, March and June)
Common work between h Reconstruction, Coalescing binaries and Burst groups
Several configurations:
“Theoretical” Virgo noise
Noise spectral density as during Virgo run E4
Noise as generated by a simulation of full Virgo in the current configuration
Include some non-stationarities (power modulations) and non-linearities (ARCH models)
Losses of lock
Main purpose: test the full In-Time chain
Cascina, July 8th, 2004

35. ‘Real Data Challenge’
Data analysis people re-starting use of real data (it had been done in the past using CITF data)
Injection of fake signals (burst and coalescing binaries) during C3 and C4
Made acting on mirrors using the coils
Will be done with an independent optical calibrator as soon as sensitivity allows it
Bring Data Analysis people closer to the real data
Cascina, July 8th, 2004

36. Off-line analysis Use of national computing centers:
- CC/In2p3 in Lyon
- CNAF in Bologna
So far mainly used to test algorithm for periodic sources search
Roma, Bologna, Napoli (about 30 machines) within INFN-Grid
Lyon (25 processors) as a classic batch system
Cascina, July 8th, 2004

37. Collaborative data analysis
LIGO-Virgo MoU proposal
“Topic driven” collaboration, focused initially on a two well defined subjects (inspirals+bursts)
Real analysis will start when a comparable sensitivity will be reached
In the meantime:
open meetings between the people involved
joint working groups will write a white paper
Responsibles for the joint working groups (Virgo side):
A. Vicere, B. Mours (inspirals)
P. Hello, L. Brocco (bursts)
The activity is coordinated by the data analysis coordinator.
Full ability of anyone in the collaboration to participate in any analysis, accordingly with the internal rules of each experiment.
Meeting at Cascina two weeks ago
Need to start an effort at an European level
Hope WG2 of ILIAS-GWA will help
Cascina, July 8th, 2004

38. Conclusions Construction of Virgo completed
- Last suspension assembly completed in November 2003
Detector commissioning started in October 2003
- Commissioning of the 3 km arms completed
- Recombined interferometer control 95% completed (noise identification starting)
- Commissioning of recycled interferometer starting
Data analysis preparation progressing
Hope to have first scientific run in 2005
Cascina, July 8th, 2004

39. IMC- ITF spurious fringes
backscattering
Input laser beam
ITF reflected beam
Backscattering on mode-cleaner mirror
 fringes at ITF input
 spurious signals in mode-cleaner control
 input mode-cleaner lock losses
Consequences:
need to misalign recycling mirror
 reduction of recycling gain
lock acquisition longer
 smaller duty cycle
fringes
Possible solutions: 1) use of Faraday isolator
2) better mode-cleaner mirror
Solution 2) adopted so far
Cascina, July 8th, 2004