LISA Interferometry Stanford University July 25th, 2006

1
LISA InterferometryStanford UniversityJuly
25th, 2006
Guido Mueller University of Florida
mueller_at_phys.ufl.edu
2
Table of Content

• Gravitational Waves
• Basics
• Sources
• LISA Mission Concept
• Gravitational Reference Sensor Focus on LTP
• Interferometry Measurement System (IMS)
• Principles
• UF Testbed
• Current Mission Status

(see S. Hughes on LISA Symposium)
3
Gravitational Waves

• What are gravitational waves?
• Generated by huge accelerated masses such
• as black holes in a binary system
• Amplitude is a relative length change h dL/L

..
2G I
h
I Quadrupole Tensor
c4 r
2G mv2
r1r2



10-17 - 10-22
c4 r
d r
4
Sources
1. Super-massive Black Hole mergers
Chandra NGC6240
2. Extreme mass ratio Inpirals (EMRIs)
3. Galactic Binaries
Credit Tod Strohmayer (GSFC)
5
SMBH merger rates

• What do we know?
• Almost all galaxies host
• a massive black hole.
• But do they merge?

• Mergers in rich cluster MS 1054-03 (z 0.83)
• Shown 16 brightest galaxies. About 20 are
  merging!
• van Dokkum et al 1999, ApJ 520,L95.

• Essentially no mergers seen in cluster MS 1358-62
  (z 0.32)
• Shown 16 brightest galaxies. No apparent mergers!

Event rate At least a few events per year!
(almost certain) (Haehnelt 1994 Menou, Haiman,
Narayanan 2001 Wyithe Loeb 2003 Islam,
Taylor, Silk 2004 Sesana et al 2004)
Scott A. Hughes, MIT
LISA VI, 19 June 2006
6
SMBH Mergers

• Mass/Redshift range
• 105Msunlt (1z)M lt 107Msun
• out to z10
• Can measure binary parameters from waveform with
  lt1 accuracy.
• Predict merger weeks in advance.
• The dream comes true event Parallel
  EM-Observations

7
Angular Resolution
LISA resolution A few 10s of arcmin
along major axis, 10 along minor axis for z1
type signals.
Full moon 30 arcminutes.
The Hubble Deep Field 144 arcseconds.
8
EMRI

• EMRI Extreme Mass Ratio Inspiral
• 1-100 MS falls into 108MS
• LISA Core Target
• Test particle case for
• gravitational waves

9
EMRI

• Predominantly Stellar mass black hole capture
• Reach z1
• Track phase over 105orbits
• Determine mass and spin of SMBH
• Determine spacetime with high precision
• Measure multipole moments of SMBH
• Relativists test particle experiment

See Barack and Cutler, PRD 69, 082005 (2004).
10
Galactic Binaries
Galactic Binaries or The other noise source

• Periodic sources
• Combined GW/EM
• observations are
• nearly guaranteed
• Main Problem
• Separation of the
• signals

Known Compact systems that radiate in LISA band
Source Gijs Nelemans
11
Galactic Binaries
Binary Confusion Signal or Noise
log h
-20

• Noise, if you are interested in SMBH or EMRIs
• Signal, if you are interested in Galactic Binary
  populations

Binary confusion noise estimate
-21
Resolvable Binaries
-22
Observation 1 yr S/N 5
-23
-4.0
-3.0
-2.0
-1.0
0.0
log f Hz
12
The Mission

• Drag-free control protects the proof masses from
  the ambient environment and reduces the
  disturbances on the proof masses from the
  spacecraft.

• Three interacting spacecraft make up the
  science instrument
• Multiple combinations of one-way measurements.

Movie
13
The Mission

• Interferometer measure
• Distances between optical
• benches on different spacecraft
• Distances between proof
• masses and their benches
• With 10 pm/rtHz accuracy

• GRS
• Needs to apply forces along non
• sensitive axis to PM to keep it centered
• Gravitational Wave Signal
• synthesized from these signals

Movie
14
GRS
How to create a freely falling proof mass?

• Shield it from all other forces!

• LISA Pathfinder (LPF)
• LISA Pathfinder is a technology demonstrator for
  the LISA GRS
• The mission will test in flight
• Inertial sensors
• Free floating test masses
• Drag Free and Attitude Control System (DFACS)
• Micro-Newton propulsion technology

15
LPF Technology

• Interferometry (not LISA like)
• Laser
• Optical Bench
• Phase Meter
• Inertial Sensor
• Proof mass
• Electrode housing
• Front end electronics
• Caging mechanism
• UV discharge system
• Vacuum System
• Micro-Newton Thrusters

Courtesy of Paul McNamara
16
LPF Mission Goal
17
Launch Sites

• Baseline launch vehicle Rockot
• Proven vehicle with heritage
• SS19 ICBM!
• Launch from Plesetsk, Russia
• Winter temperature -30oC

• Target launch vehicle is VEGA
• New launcher
• ESA directive to target European launchers
• LPF could be first flight!
• Launch from Kourou,
• Winter temperature 28oC!

18
Interferometry

• Req. Sensitivity 10pm/rtHz
• Ideal Interferometer
• One laser, fixed beam splitter
• Mirrors at the far ends
• Equal Arm length
• Insensitive to frequency noise

• LISA Interferometer
• Two laser, no beam splitter
• Transponder at far ends
• Unequal Arm length
• Laser frequency noise dominates

19
Synthetic Beam splitter
Bench A PM1 f1(t) - f2(t) fibernoise Bench
B PM2 f1(t) - f2(t) - fibernoise PM1 PM2
2 f1(t) - f2(t)
To far S/C
Phase Meter 2
Phase Meter 1
Fiber to/from Second Bench

• Independent of fiber noise
• Used to phase lock local lasers
• Allows to compare both
• Interferometer arms
• Like having a beam splitter in
• a Michelson Interferometer

To far S/C
20
Unequal Arms

• Laser frequency stabilization
• Time Delay Interferometry (TDI)

21
Laser frequency stabilization
Baseline Stabilize to ultra-stable reference
cavity

• Two lasers independently stabilized to two
  frequency references.
• References 2 Zerodur spacers with optically
  contacted mirrors in ultra-stable vacuum chamber
• Pound Drever Hall stabilization scheme

22
Laser frequency stabilization

• Similar Results
• AEI Hanover
• GSFC

23
Arm Locking
Basic Idea Lock laser frequency to LISA arm
Far S/C Transponder (phase locked laser)
!
S(t) f(t-2t)-f(t) 0

• Transfer function is zero at Fourier
• frequencies fN N/2t
• Requires tailored feedback gain (1/sqrt(f))
• at and above f1
• Laser frequency noise suppressed
• at all frequencies except at fN N/2t

24
Arm Locking
Different potential realizations
Single
Common
Sagnac
Round-trip arm length
Difference between arms
Sagnac effect (rotation)
25
Unequal Arms

• Laser frequency stabilization
• Time Delay Interferometry (TDI)

First Generation X-combination
Sb(t) - Sg(t) - Sb(t- 2tg) Sg(t- 2tb)

• Requires to know the light travel times betw. S/C
  
• Ranging with 30m accuracy

Synthetic equal arm Interferometer!
26
Interferometry

• Synthetic Beamsplitter
• Ranging and TDI create a nearly equal arm
  interferometer
• Cavity and/or Arm-locking provide the necessary
  frequency stabilization
• Still Need
• Phasemeter?
• Ground testing w/o having 5Gm real estate?

27
Phasemeter

• Requirements
• 2-20 MHz signal frequencies, changing by several
  MHz (Doppler)
• Frequency noise of 30Hz/Hz1/2 (cavity-stabilized
  lasers)
• 10-5 cycles/Hz1/2 residual noise (LISA
  requirement)

28
Phasemeter (JPL)
Tracks the Phase of RF signal with NCO
Devil is in the details (Downsampling,
anti-aliasing filters, timing accuracy, etc.)
I/Q demodulation with tracking NCO
29
Phasemeter JPL
Equivalent Optical Setup

• Digitally tested dynamic range requirement.
• Digitally generated 3 independent, laser-like
  noise sources such that,
• Phase 0 Phase 1 - Phase 2 0

(Results from Daniel Shaddock, Brent Ware, Bob
Spero, JPL)
30
UF Phasemeter
Started with a Software Phasemeter
Characteristics

• 80kHz sampling
• 10 kHz signals
• PM is run off-line

Applications

• Verification of Algorithm Analytic Models
• Initial simulator experiments

JPL-type phasemeter near completion at UF
31
Electronic Phase Delay
How can we recreate the long LISA arm on Earth
(16 sec)?
Just delay the phase of the laser.

• demodulate it with a stable oscillator (another
  laser)
• digitize the difference phase
• store it in a buffer
• regenerate the signal

32
Electronic Phase Delay

• UF technique
• Laser Phase replaced by beat note phase
• Beat note phase delayed electronically (EPD).
• LISA photodiodes replaced by electronic mixers.

LISA
33
Evolution of EPD System
Max. Signal
System

Date

Hardware

Chan.

Max. Delay

Freq.

Summer
200 kHz PCI
Original

30

kHz

2

80
s

2004

card

Summer
Current

Pentek

5

MHz

4

6
s

2005

Fall

Pentek w/
Future

2
0

MHz

4

35
s


2006

PMs NCOs


Depends on resolution BW
34
TDI-Tests at UF
First experimental verification of TDI!
Rachel Cruz, Michael Hartman, UF
35
Two-Arm Experiment
(Nearly) full scale LISA signal Limited by
Transponder Noise
36
Experimental Result

• Results currently limited by PLL performance

5 orders suppression
Phase Noise cycles/rt(Hz)
Rachel Cruz
Frequency Hz
37
Main Next Step
Build final (LISA-like) phasemeter
Max. Signal
System

Date

Hardware

Chan.

Max. Delay

Freq.

Fall

Pentek w/
Future

2
0

MHz

4

35
s


2006

PMs NCOs

38
Optical Layout
Reference laser
Master laser
LISA Simulator with 1 Laser on each S/C.
39
LISA Signals
S21(t)
S23(t)
Reference laser
S13(t)
S12(t)
S32(t)
Master laser
S12(t) f20(t-t21)-f10(t)
S31(t)
40
TDI with Phase lock loop
S21(t)0
S23(t)
Reference laser
S13(t)
S12(t)
S32(t)
Master laser
S12(t) f20(t-t21)-f10(t)
S31(t)0
41
Optical Layout

• Note All optical path are common mode
• Insensitive to Optical path length changes!

42
Doppler and Signals
PD

• Current Setup
• Cancels all optical path length changes.
• No GW-signals or Doppler shifts

PD

• Future Setup
• Split Optical Path
• Doppler shift can be added
• in the EPD unit
• GW-signal can be added via PZT
• Sensitive to acoustic noise
• Will be moved in Vacuum

PZT
PD
PD
43
Arm-Locking
S21(t)0
S23(t)
Reference laser
S13(t)
S12(t)
S32(t)
Master laser

• Common Arm-locking
• Sagnac Arm-locking

S31(t)0
44
Arm Locking
Intermediate Arm locking Experiment Understand
and develop feedback system in a real
electro-optical setup
45
Arm Locking

• Compare to LISA

46
Arm Locking

• EPD using 25 MHz digitization rate, delay of
  1.065ms,

47
Arm Locking

• Out-of-loop
• Primary beat note demodulated to 10kHz
• Phase of 10kHz signal measured using software
  phase meter.

Ira Thorpe
48
Arm Locking

• Out-of-loop measurement of primary beat note
  using frequency counter.

Ira Thorpe
49
Arm Locking

• Out-of-loop measurement of primary beat note
  using frequency counter.

Ira Thorpe
50
Summary

• LISA
• Was considered a very challenging mission
• No ground testing possible
• No technology heritage for any of the major
  technologies
• GRS
• Interferometry
• Data Analysis

51
Summary

• LISA
• GRS
• Will be flight tested in LTP around 2009/10
• LTP ground tests look very promising so far
• Interferometry
• Basic concepts of TDI, Arm-locking, clock noise
  removal are well understood
• Experimental tests at component level are
  progressing very well
• EPD unit enables detailed ground testing of
  TDI/AL

52
Summary

• LISA
• Remaining Challenges
• How to move the telescope w/o distorting the
  measurements?
• Do we need to measure these distortions and
  correct for them?
• How to align the spacecraft to acquire lock?
• Stable materials and components
• Laser switch, Fiber launcher, Vacuum system,
  Discharging, PAA actuator,
• Data Analysis
• Some signals might cover other signals some
  people are more interested in.
• Does this sound different from other missions?

53
Short LISA History

• Foundation paper in 1984 by Bender, Faller, Hall,
  Hils and Vincent
• Concept developed through
• Concept studies 84-93
• ESA Pre-Phase studies 93-98 (cf., PPA2
  document)
• NASA Team-X study 98
• ESA Industrial Phase A Study 98-00 (cf., FTR
  and STS documents)
• GSFC Project Office formed in 01, technology
  planning and development commenced.
• Flight demonstrations (LISA Pathfinder and ST-7)
  initiated in 00-01
• NASA Formulation Phase began Oct. 04
• ESA Industrial Formulation Study begun at
  Astrium/Friedrichshafen Jan. 05, finished Phase
  I in Oct. 05
• Concept has not significantly changed since PPA2
  in 1998.
• Current focus
• Architecture definition and refinement, design
  trade studies
• Technology development
• LISA Pathfinder and ST-7

We entered Phase A late 2004!
Slide stolen from Robin Tuck Stebbins LISA
Symposium Talk
54
Mission Status

• ST7 brings the least well-tested LISA
  instrumentation, DRS, to TRL level 9
• Preparations for 2010 launch will already
  greatly enhance
• Experience in building flight models
• Experience in tightly-coupled NASA/ESA
  cooperation
• Results from 2010 launch will be in time to
  inform formulation

FY07
FY08
FY09
FY10
FY11
launch
HW delivery
Phase C/D
Phase E
Phase A (survival)
Phase A
Phase B
Slide stolen from Colleen Hartman, LISA Symposium
55
Mission Status

• Budget requirements have necessitated Beyond
  Einstein be sequential missions rather than
  parallel efforts
• Funding wedge for first BE mission start in 2009
• One of 3 will go first LISA, Con-X, JDEM
• Special BE NRC panel in 2008-9

Instead of two parallel lines of sequential
missions
We hear you
JDEM Additional competition!
From Colleen Hartman, LISA Symposium
56
LISA Project

• ESA/EU
• ESA/Estec
• Astrium, Germany
• AEI Hanover
• University Trento
• University of Birmingham
• University of Glasgow

• NASA/US
• GSFC
• JPL
• University of Florida
• JILA
• Stanford
• University of Washington

The End
57
Optical Bench
Phase Meter 2
Phase Meter 1
Phase Meter 3
Fiber to/from Second Bench
to/from far SC
PM
from Laser Bench
58
Optical Bench
Phase Meter 1

• Bench A
• PM1A f1(t) - f2(t) fibernoise
• Bench B
• PM1B f1(t) - f2(t) - fibernoise
• PM1A PM1B 2 f1(t) - f2(t)
• Independent of fiber noise
• Used to phase lock local lasers
• Allows to compare both Interferometer arms
• Like having a beam splitter in a Michelson
  Interferometer

Fiber to/from Second Bench
from Laser Bench
Only works if OPL in fiber is independent of
propagation direction!
59
Optical Bench
Polarization Sagnac Interferometer for Optical
Fiber Tests at UF
Fiber
Pol
l/2
l/4
Laser
l/4
BS
Pol
Parallel tests in Glasgow, Hanover
60
Optical Bench
Phase Meter 2
Phase Meter 1
Phase Meter 3
Fiber to/from Second Bench
to/from far SC
PM
from Laser Bench
PM 2 PM 1 Distance PM - SC
61
Optical Bench
Phase Meter 2
Phase Meter 1
Phase Meter 3
Fiber to/from Second Bench
to/from far SC
PM
from Laser Bench
PM 3 Distance SC SC How?
62
Optical Bench

• Phase Meter 3 on S/C 2 and 3
• Used to Phase lock local laser

Phase Meter 3
PM 3A PM 1A
To Laser frequency actuator PLL
to/from far SC
PM
from Laser Bench
63
LISA
Master S/C
Slaved S/C
Slaved S/C
64
Optical Bench

• Phase Meter 3 on Master S/C 1

Phase Meter 3
PM 1A PM 3A

• f1(t)-f1(t-2t1)GW1
• (Unequal Arm MI)
• Dominated by Laser frequency noise df
• 1000 cycl./rtHz noise

PM
from Laser Bench
to/from far SC
65
Primary Hardware

• Key Features
• 4 Channels
• 14-bit ADC
• 16-bit DAC
• 1 GB SDRAM
• 100 MHz sampling
• 5 FPGAs
• PowerPC processor
• Ethernet, serial, VME

66
Primary Hardware

• Key Features
• 4 Channels
• 14-bit ADC
• 16-bit DAC
• 1 GB SDRAM
• 100 MHz sampling
• 5 FPGAs
• PowerPC processor
• Ethernet, serial, VME

67
Gravitational Waves
NS/NS merger (MNS 3x1030kg 1.4 MSun)
1. Smallest Distance dmin 20km (2xDiameter
of NS)
2. Potential Energy E - GM2/d 3x1046J
3. Newton f (d100km) 100 Hz, f (d20km)
1 kHz
4. Takes about 1s to get from 100km to 20km
5. During that second nearly half of the
Potential Energy is radiated away!
6. Assume binary is in the Virgo cluster (15 Mpc
6x1024 m)
We receive about P1..100mW/m2 from each
binary! Like full moon during a clear night!
68
Gravitational Waves
We can see the moon, why havent we seen
Gravitational Waves yet?
GW-Amplitude hdL/L is
G/c4 10-45s2/kg m
69
Gravitational Waves
We can see the moon, why havent we seen
Gravitational Waves yet?
GW-Amplitude hdL/L is
Answer Space is stiff
G/c4 10-45s2/kg m
Our example (f400Hz)
Or 1am over 1km
70
LISA

• LISA will probe space and time
• at the forming edges of black holes listening
  to the sounds of vibrating spacetime
• the booming roar of supermassive black holes
  merging
• the chorus of death cries from stars on close
  orbits around black holes
• and the ripping noise of zipping singularities

Even the NASA-folks were a little excited about
LISA
Unfortunately, LISA will be unmanned and not on
Mars
Copied from Beyond Einstein from the big bang
to black holes
71
UF Benchtop

• Ground-based Simulator
• First Generation of Experiments
• Frequency-stabilized lasers
• Arm-locking
• Time Delay Interferometry (TDI)
• Future Experiments
• Doppler shifts
• Clock noise, laser com.
• GW-signals

Long Term Goal Provide realistic data streams
with injected GW signals
72
The Mission
Current Design of single tube
73
GRS-Challenges

• A few (obvious) forces pushing the PM
• Lorentz Force
• Charged PM moving in variable solar magnetic
  field
• Charge Control (UV-light, continuous or every
  10-20h?)
• Magnetic Force
• Magnetic Susceptibility couples to magnetic
  fields
• Gold Platinum Alloy cm 0 (Problem Grains in
  PM have variable cm)
• Self-Gravity from S/C
• 1kg mass 10cm from PM gives a gradient of
  10-7m/s2/m
• S/C motion lt 10nm/rHz (Design of S/C,
  mN-Thrusters)

74
GRS

• A few (not so obvious) forces pushing the PM
• Patch Fields
• Crystal Boundaries create voltage potentials
• Gas pressure noise
• Gas hitting the PM from both sides
• mDa PDT requires DT lt 10-4K/rHz and P
  lt 10-8torr
• Thermal photon pressure
• Black Body Radiation from walls
• mDa DT requires DT lt 10-4K/rHz

75
Timing Error
The delay time of the EPD, just as the optical
delay time of the LISA arm, will not fall exactly
at one of the sampling points of the data stream.
Define the timing error as
76
Suppression Limit

• The timing error in the experiment lt ?tmax ½
  tsamp 6.25 µsec
• Interpolation can be used to reduce the timing
  error
• Experimental results appear to hit another noise
  source at 5x10-5 cycles/rt(Hz)

Phase Noise cycles/rt(Hz)
Smin(f,?tmax)
Exp. Time-delayed Comb.
Frequency Hz
77
Two-Arm Experiment
78
Benchtop Simulator
79
Data Analysis Challenge

• The signals
• 3 Hz sampling of 18 beat signals
• Time Delay Interferometry (TDI) algorithms to
  remove laser and clock frequency noise
• Auxiliary sciencekeeping data (solve for PM
  motion)
• More than 10,000 interfering GW signals.
• Signals have to be
• Identified, separated, tracked, and subtracted
  from data stream
• Source direction can be determined
• Frequency and amplitude modulation from orbital
  Doppler shifts
• Phase modulation from time-of-flight across
  antenna