Transcript 061121-atf2-suehara

Status of fringe stabilization
of Shintake-monitor
Taikan SUEHARA
The University of Tokyo
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Beam Size Resolution
vs.
Fringe Phase Fluctuation
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Required Modulation Resolution
GOAL : to measure 35 nm beam size
by < 2nm resolution
zoom
Beam size
modulation
33nm
73%
35nm
70%
37nm
67%
2nm beam size corresponds
to 3% modulation (around 35nm).
We need 3% modulation resolution.
Then, how much stability of laser fringe phase
do we need to achieve 3% modulation resolution?
 We performed a Monte Carlo simulation to obtain that relation.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Modulation Error Estimation
Assumed measuring condition for simulation is:
45 points (phases) meas., 1 bunch for each point
It’s determined by desired measuring time (1 minute)
45 points  1 bunch + same for background reduction = 90 bunches.
90 bunches / 1.5 Hz (ATF2 operation freq.) ~ 60sec.
Simulation method is shown below (by example)
h
A sample for 50% modulation, 0.2 rad (s) phase jitter
count
intensity
Graph
Original curve
Fitted curve
0.7
h
A sample histgram of 10000 fitting results
50% modulation 0.2 rad phase jitter
s
of
gaussian fitting corresponds to
400
the resolution of modulation.
Entries
Mean
RMS
10000
0.4906
0.01182
350
300
0.6
250
0.5
200
150
0.4
100
0.3
50
0
1
2
3
4
5
6
0
0
phase [rad.]
0.1
0.2
0.3
0.4
0.5
0.6
fitted modulation
A sample of error simulation caused by phase jitter
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Required Fringe Stability
The result of simulation:
• modulation uncertainty
is proportional to
phase jitter.
• About 30nm stability
corresponds to 3%
(goal) modulation
uncertainty.
• But we should consider
another errors.
We keep safe factor 3.
As a result,
We should develop a 10nm level fringe phase stabilization system.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Phase Stabilization Status
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
The Method of Phase Detection
Microscope lens
Image sensor
Fringe magnification by a lens with a linear image sensor
• The laser beams pass
through the microscope
lens to be magnified,
and create fringe pattern.
(lens works like a double slit)
• The phase of the fringe
corresponds to relative
phase of 2 laser beams.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Location of Monitors and a Scanner
Monitor and scanner location
Phase scanner
Beam sampler
off-axis
monitor
(ch2)
IP
(on-axis)
off-axis
monitor
(ch1)
• We cannot place the phase
monitor on IP. We place
it “off-axis” position.
• To cancel out difference
of the phase between
IP and monitor position,
we place the same monitor
on the other side (ch2).
• We place a phase scanner
(delay line with piezo mover)
on one side.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Setup
Linear image sensor
(photo by HPK website)
256 pix, 25um pitch
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Fringe Phase Scanning System
• Optical delay line used
for phase stabilization
and scan
• Resolution of piezo
stage is 1nm,
corresponds to 2nm
phase resolution.
• The stage can be
controlled by PC up to
300Hz (sufficient speed
for stabilization).
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Linear Image Sensor Data
h
Entries
Mean
RMS
h2
256
126.7
73.71
power
intensity
h
43000
42000
h1
2pi : effect
of baseline
7
Entries
Mean
RMS
10
10000
4.379
2.915
106
interferometric mirrored
peak
peak
41000
105
40000

39000
104
38000
103
37000
36000
102
0
50
100
150
200
250
Sensor channel [1pix = 25um]
Raw spectrum. Wave of dense pitch
0
1
2
3
4
5
6
7
8
9
10
fourier frequency [2pi = 1pix. cycle]
Fourier spectrum. Clear peak near 2.5.
• We use Fourier transform to obtain phase (at central channel)
for high resolution and noise reduction.
• Clear peak can be seen near 2.5 on Fourier spectrum.
It corresponds to the wave seen on raw spectrum.
• We use the Fourier phase (i.e. argument of complex Fourier)
of the peak for the phase stabilization.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Preliminary Result(1)
Simply measured phase for 50 sec. No stabilization.
ch1 (FT peak period)
ch1 (fixed period)
ch2 (FT peak period)
ch2 (fixed period)
phase(rad.)
4
3
2
1
0
-1 0
5
10
15
20
25
30
35
40
45
50
-2
tim e (sec.)
FT peak period: the phase of the Fourier peak freq.
Fixed period: the phase of fixed freq. (near the peak).
• Ch1 and Ch2 are almost opposite because
they face opposite directions (it’s expected behavior).
• This shows correlation of phases at two monitors.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Preliminary Result(2)
unstabilized
ch1
ch1
ch2
ch2
2
ch1
ch1
ch2
ch2
2
(FT peak period)
(fixed period)
(FT peak period)
(fixed period)
1.5
1
1
0.5
0
0
500
1000
1500
2000
-0.5
2500
3000
3500
4000
measured phase (rad.)
measured phase (rad.)
1.5
stabilized (ch1-fix)
(FT peak period)
(fixed period)
(FT peak period)
(fixed period)
0.5
0
0
-1
-1
-1.5
-1.5
-2
-2
tim e (sec.)
•
•
•
•
500
1000
1500
2000
2500
3000
3500
4000
-0.5
time (sec.)
Stabilization effect is clearly observed.
s = 0.076 (3.2nm) for ch1 fixed (stabilized channel)
s = 0.178 (7.5nm) for ch2 fixed
(unstabilized channel, except long-time drift)
Almost achieved 10nm stability for very bad condition
(no cover, lenses with rods). We can improve further.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Phase Monitor Errors
(by Position / Angle Jitter)
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Basic Check of Fringe
Characteristics
5
• Crossing angle & Fringe pitch
• Lens mag. & Fringe pitch
– Lens magnification is inverseproportional to fringe pitch
4
double slit distance [mm]
– Crossing angle is proportional
to “slit distance” (i.e. inverseproportional to fringe pitch)
4.5
3.5
y = 7.6082x + 0.0471
3
2.5
2
1.5
系列1
線形 (系列1)
1
0.5
0
0
0.2
0.4
0.6
crossing angle [rad.]
0.8
• Lens NA & Acceptable crossing angle
– Clear view until NA
– Partial view until a little wider than NA
– 128 deg. for NA=0.9, 142 deg. for NA=0.95
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Resolution of Fourier Transform
• With random noise (S/N > 1)
• Rotate phase, and check maximum error of
detected Fourier freq. & phase
• Freq. error is
0.1% @ S/N=1
• Phase error
(fixed freq.)
is 0.1rad.
1
0
0.2
0.4
0.6
0.8
1
1.2
最大誤差
0.1
0.01
干渉縞周期
位相(ピーク)
位相(固定)
0.001
0.0001
S/N比
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Calculation of Jitter Requirement
• I calculated the jitter
object
axis
freq.
phase
1um
requirement for Fourier- detector XZ
100um
angle
45mrad
limit freq & phase reso.
X
1um
Z
100um
• 1um position stability lens
angle
45mrad
10urad
is required for detector
X
beam
angle
600urad 11urad
and lens (X axis)
• 10urad angle stability freq. = acceptable fluc. for 0.1% reso.
phase = acceptable fluc. for 0.1rad reso.
is required for lens
Fringe pitch: 50um
and beam.
• Going to check the jitters by vibration monitor.
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Other Issues
• IPBPM & Shintake-monitor
– IPBPM with laser port?
– How to align laser beam without a slit?
Or the slit is available with IPBPM?
– Doesn’t the laser beam perturbate IPBPM
signal?
• We must prevent the air flow
– Temporal cover with acryl board & alumi frame
– Black curtain (also for safety of high power
laser)
• Image sensor with smaller
pitch
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
Summary
• To achieve 2nm resolution, we have to
obtain
10nm fringe stability.
• First fringe stabilization result shows < 10nm
fringe fluctuation.
(because of high reso. by Fourier transform)
• The requirement for position/angle stability
of the monitor and the lens is not so hard.
(1um,10urad level)
• The stabilization of this method is hopeful!
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22
以上。
Taikan SUEHARA, ATF2 meeting, KEK, 2006/11/22