Paolo La Penna The New Virgo Injection Bench installationN5-WP1 7 TH MEETING, Hannover, 12/12/2005 The installation of the new VIRGO Injection Bench Paolo.

Paolo La PennaThe New Virgo Injection Bench installationN5-WP1 7TH MEETING, Hannover, 12/12/2005

The installation of the new VIRGO Injection Bench

Paolo La Pennafor the VIRGO collaboration

European Gravitational Observatory


Recombined configuration: first half of 2004

Nd :YAG= 1064 nm

20 W

DETEC TIO NSYSTEM

WIp la ne

R= 0.88

NIp la ne

R= 0.88

WER = 3500 m

R= 0.99995c urv

NER = 3500

R= 0.99995c urv

PRp la ne

R= 0.925

BS

INPUTM O DE

C LEANER(144 m )

6 m 6.4 m

5.6

m

3 km

3 km

O M C

M o d ula to rsing le fre q ue nc y

6.26 M hz

Nd :YAG= 1064 nm

20 W

DETEC TIO NSYSTEM

WIp la ne

R= 0.88

NIp la ne

R= 0.88

WER = 3500 m

R= 0.99995c urv

NER = 3500

R= 0.99995c urv

PRp la ne

R= 0.925

BS

INPUTM O DE

C LEANER(144 m )

6 m 6.4 m

5.6

m

3 km

3 km

O M C


6.26 M hz

IB


Recycled interferometer: July 2004

Nd :YAG= 1064 nm

20 W

DETEC TIO NSYSTEM

WIp la ne

R= 0.88

NIp la ne

R= 0.88

WER = 3500 m

R= 0.99995c urv

NER = 3500

R= 0.99995c urv

PRp la ne

R= 0.925

BS

INPUTM O DE

C LEANER(144 m )

6 m 6.4 m

5.6

m

3 km

3 km

O M C


6.26 M hz

Nd :YAG= 1064 nm

20 W

DETEC TIO NSYSTEM

WIp la ne

R= 0.88

NIp la ne

R= 0.88

WER = 3500 m

R= 0.99995c urv

NER = 3500

R= 0.99995c urv

PRp la ne

R= 0.925

BS

INPUTM O DE

C LEANER(144 m )

6 m 6.4 m

5.6

m

3 km

3 km

O M C


6.26 M hz

IB


PR reflection inside IMC: frequency noise

Black: PR misaligned

Red: PR aligned


PR reflection scattering inside IMC

MC scattering (10 ppm)

PR

IMC

Cavity effect

(10% fringes)


Plane Power Recycling mirror

The old PR was made by two parts: a curved one inside an external cylindrical glass mass (it was a lens, part of the telescope for collimating inside the ITF)

The curved part was fixed to the cylindrical by means of a steel ring and pression screws;

There was evidence of mechanical resonances (drifting in time): problems in locking acquisition and in the future Frequency Stabilization: need of a monolitic mirror

Decided to make it plane (get rid of transverse movements of the beam induced by the suspension displacements)


IMC and RFC alignment

10% ABP OFF

ABPdismounted

The beam was atomatically aligned on the RFC (fixed below the IB) with ABP

The IMC was partly automatically aligned (feedback only on the MC mirror

The RFC was more stable than IMC:

the AA of the RFC worked quite well

the poor AA of the IMC couldn’t correct the drifts


PR reflection: Input Beam Attenuation (temporary solution)

From the laser

To the interferometer

10% M6 IMC

Reference Cavity700 mW (10% of the

full power)


The PR feedback inside the IMC (mid of 2004)

Simulations on a lock acquisition technique developed following the LIGO experience

Locking trials with this baseline technique failed (first half of July)

Beam attenuation installed (summer)


HR M6 vs attenuated M6

R=10% M6HR M6

HzHz

(Same y-scale)


The PR feedback inside the IMC (end of 2004)

Restart of the locking trials with the baseline technique (21st September)

Establishement of theVariable Finesse lock acquisition technique (October)

PR locking acquisition after one month (end of October 2004)

CONFIRMATION OF THE BACKSCATTERING PROBLEM

NEED OF AN OPTICAL ISOLATOR (FARADAY)


Resume: why to build a new Injection Bench

The reflection of the light inside the Input Mode Cleaner increased the frequency noise, thus making it impossible to lock the recycling cavity

The use of a Faraday isolator was necessary There was no way to accomodate a Faraday isolator and

respective telescope on the old IB: the bench had to be replaced

In addition to this:

It was decided to replace the curved PR with a plane PR (less coupling with the transverse PR displacement): a short (parabolic) telescope on the IB was necessary

It was decided to change the automatic alignment system of the injection system: separate alignment of the Input Mode Cleaner and Reference Cavity


How did we proceed

The aforementioned problems were individuated in many data takings (C1-C4)

The mechanical design of the bench was performed using essentially Autocad and Ansys

The optical simulation of the bench was performed using mainly Zemax (simulation of optical aberrations and stray light), and by Optocad

The Faraday isolator was tested in Nice with the 20W laser

The new IB was mounted and prealigned in Class 10 room (after old IB dismounting)

Once prealigned it has been inserted into the tower


Upper part layout

Autocad drawing Optocad drawing


Lower (RFC) part layout

Autocad drawing Optocad drawing


Autocad 3-D simulation (example)


ANSYS model

Reference

Cavity

Marionette


ANSYS simulation

Uy Tx

90807060504030

Tz

Susp. Wire

couplers

ProbablyDihedron

Legs(measured in

september 2005)


Plane PR: Beam simulation with spherical mirrors

Input mirrors End mirrors


Plane PR: Beam simulation with parabolic mirrors

Input mirrors End mirrors


Plane PR: (new) 6 parabolic off-axis telescope

IMC: waist 4.9 mmCondensing telescope

FI: waist 2.65 mm

M5: f= 75 mm

M6: f = 600 mm Ø 10 cm

to PR: waist=20 mm

d= 675 mm

With plane PR: a big magnification is needed, the telescope has to be short (about 700 mm)Parabolic mirrors are needed

The computed curvature radius are not off-shelf: custom mirrors (Optical Surface)


Parabolic telescope mechanical mount

Three closed loop picomotors

Two open loop picomotors

X and Z closed loop translators

Mounts designed by ourselves and

Machined by a workshop close to Virgo

Mounting commercial actuators


Alignment of the parabolic telescope (Class 10 room)

Laser


Starting conditions: misaligned, mismatched

AUTOCOLLIMATOR


1st autocollimator alignment

AUTOCOLLIMATOR REF. MIRROR

AUTOCOLLIMATOR


2nd autocollimator alignment

AUTOCOLLIMATOR

REF. MIRROR

AUTOCOLLIMATOR


Autocollimators aligned

AUTOCOLLIMATOR

AUTOCOLLIMATOR


1st mirror alignment

AUTOCOLLIMATOR

AUTOCOLLIMATOR


1st mount alignment

AUTOCOLLIMATOR

AUTOCOLLIMATOR


2nd mount alignment

AUTOCOLLIMATOR

AUTOCOLLIMATOR


2nd mount alignment

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Mirror matching

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Focussing

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Off-axis

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Off-axis

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Off-axis

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Autocollimation

AUTOCOLLIMATOR

AUTOCOLLIMATOR


Beam dumpers: aluminum boxes and buffle-glass plates


Beam dumpers and PSDs

PSD

PSD

PSD

PSD

Beam dumpers


Resume of the main differences between new and old IB

Different shape and strutcture of the bench (octagonal, larger, M6 grid of holes, …)

Thinner suspension wires Presence of the Faraday isolator Presence of a collimating (reducing) telescope between the

Input Mode Cleaner (two lenses) Presence of a short (less than 1-m-long) off-axis parabolic

telescope (×8) Separate alignment of the IMC and the RFC RFC placed after the IMC (no beam on the RFC when the IMC

is not locked): it should make the beam on the RFC more stable Use of several PSDs to monitor the beam on the bench Use of larger mirrors Use of specifically designed beam dumpers


Beam diameter 10 mm Beam diameter 40 mm


Actions after the recycling cavity locking

Decision to revise the Injection Bench design (June 2004)

Creation of a group charged to redesign the IB (July 2004)

Choice of the Faraday isolator and simulation of several possible telescopes (autumn 2004)

Decision to install a plane PR (and therefore an off-axis parabolic telescope) (November 2004)

Develope of the IB optical and mechanical design and starting of orders and (first months of 2005)

Arrival of the main part of the components (May 2005)

Arrival of the bench (July 2005)

Arrival (delayed) of the last parabolic mirror (M5, 75 mm focal length) (August 2005)


Actions after the recycling cavity locking

Dismounting of the old IB and starting of the mounting of the new IB (September 2005)

Insertion of the new IB into the IB tower (begin of november 2005)

IB suspension tuning (november 2005)

Bench prealignment and local controls tuning (end of November 2005)

Mounting of the plane PR (end of November 2005)

Local control of the bench (December 2005)

Nd:YAG injection and beam alignment (December 2005)

Alignment of the IB optics (now)


Brewster dielectric polarizers

Rotator

February 2005


Mid September 2005


Begin of October 2005


Main activities before inserting the bench into the tower

Change of the input flange (larger windows) Change of the Brewster link (to accomodate a larger

beam) Suspension tuning (filter 0 and 7 blades): the bench

is heavier (about 150 kg) Adjustment of the wires length Full tuning and balancing of the suspension Mounting of the coils: close to the center of mass

plane (more stable equilibrium of the bench) Cabling (of coils and bench actuators)


Insertion into the tower (begin November)

Detector table


Insertion into the tower (begin November)

Detector table


Begin of November 2005


Now


Now


Main activities after inserting the bench into the tower

Bench heigth setting: some difficulties with the suspension point motors

Balancing of the bench Orientation of the bench with an HeNe coming from

the MC tower: some difficulties in the filter 7 rotation range

Test of the coils: some problems in the frames interrupting RFC beams

Orientation of the local controls target: quite difficult because the small angular range in its orientation


Local controls

Beam diameter 10 mm Beam diameter 40 mm

The IB local controls use a target, placed in front of the dihedron,

with reference markers (bench translations) and reflecting HeNe beams (angles)

The setup had not been changed with respect to the old IB


Old IB local controls



New IB local controls

After the realignment of the reference beams and target:

Closing of the coarse and fine with the old IB correction filters and essentially the same gains (end of November)

Adjustment of the reference spots Fine alignment of the bench with the MC HeNe (begin of

December) Measurement of the mechanical TF (last week)


Preliminary: New TF measurement

(Henrich Heitmann and Paolo Ruggi)


Full alignment procedure

After this first alignment in Class 100 room, the bench has been mounted inside the tower.

The dummy dihedron is mounted, the LC target is realigned, the local controls of the bench are restarted.

The IB is aligned using the HeNe coming from the MC tower

The Nd:YAG beam is injected into the bench and aligned until a light spot is visible on the MC mirror

The optics on the bench is realigned using the IB actuators

The true dihedron is reinstalled, the IB is put in vacuum, the IMC aligned and locked

After IMC locking, it should be possible to make the final tuning and alignment using the bench actuators.


Next steps

Present status:

The fake dihedron is still mounted on the bench The Nd:YAG beam (up to 5 W for the moment) has been sent

into the bench: some light (with even some interference) is visible at the end of the MC

The optics on the bench are being realigned using the IB actuators: we perform this operation staying inside the tower, with the laminar flux on and the LC closed (they are very robust)

The true dihedron will be reinstalled this week, then the IMC will be aligned (we hope it will be possible in air, in order to be able check the RFC alignment)

the IB will put in vacuum, the IMC aligned and locked

After IMC locking, it should be possible to make the final tuning and alignment using the bench actuators.

The beam will be aligned in the ITF

Paolo La Penna The New Virgo Injection Bench installationN5-WP1 7 TH MEETING, Hannover, 12/12/2005 The installation of the new VIRGO Injection Bench Paolo.

Documents

new injection bench

th meeting

new ib

ib slide

injection system

power slide

summer slide

tower slide