ANATAC Meeting

1
ANATAC Meeting

• Line Length Correction Status
• 2004-Apr-23

2

• In November,
• Tests and theory indicated that phase change
  occured when the two lightwave polarizations were
  not fully aligned when the fiber (antenna) moves
• Further tests indicated that
• Misalignment of the polarizations can be added
  by either fiber or other components
• Effect of fiber optic circulators on
  polarization alignment is large
• Effect of 25km of fiber is fairly small

3

• Optical Circulator - Polarization Independent
  Input Type
• From US Patent 4,650,289 Mar 17, 1987
• A ? B ? C ? D
• Any path length difference on the split-path
  causes a wavelength dependent polarization change

4
(No Transcript)
5

• Polarization Requirement
• Depends on relative polarization alignment
• Misalignment must be minimized
• Alignment must be done at the central building,
  before the two lightwaves are combined
• Fiber introduces a misalignment that is
  unavoidable in the present baseline
• Maximum angular misalignment introduced was
  measured for 25 km benchtop spool, about
  equivalent to 15 deg of arc on the Poincare
  sphere, or 97 alignment.
• This is enough to cause a 3 amplitude reduction
  (compared to perfect alignment)
• The phase change induced (with no circulators)
  was below 0.5 deg at 20 GHz, the limit of the
  measurement (about 100 fsec)
• Since the spec is 12.2 fsec, we have not retired
  the risk, we need to measure to better precision
• A reason to be optimistic the phase change was
  unmeasurable for a 300 degree rotation of a
  fairly crude azimuth mock wrap. This is a much
  worse case than for ALMA, which will have a
  carefully engineered wrap and will in general
  make much smaller angular moves.

6
New Block Diagram with Optical Circulators
Removed
Line Length Corrector (64 ea.)
At Each Antenna
50 MHz
Master laser stability 1e-11 short term (gt30km
coherence) 1e-10 long term
?
Separate photodetector for each band, output in
waveguide.
PD
Laser Synthesizer (4 ea.)
Loop Filter
Master (stabilized) Laser
1556.21 nm
Reference out 27 to 142 GHz
PD
Fiber Stretcher
up to 14 km of fiber
PBS
Slave (tunable) laser
OFS
Faraday Mirror
PD
1556.41557.4 nm
Harmonic Mixer
Loop Filter
?
25 MHz
Not shown optical signal distribution and
switching.
9 to 11 GHz
50 MHz
PD photodetector PBS polarization beam
splitter OFS optical frequency shifter
optical signals electrical signals
7
References 1 L. Ma, P. Jungner, J. Ye, J.L.
Hall, , Delivering the same Optical Frequency at
two places accurate cancellation of phase noise
introduced by an optical fiber or other
time-varying path, Optics Letters, Vol. 19, No.
21, Nov. 1, 1994, pp. 1777-1779 2 M.
Martinelli, A Universal Compensator for
Polarization Changes Induced By Birefringence on
a Retracing Beam, Optics Communication, Vol. 72,
No. 6, 15 Aug 1989, pp. 341-344
8

• Prototype Master laser was delivered and accepted
• Coherence was adequate to close the loop on a 5
  km spool of fiber. Longer lengths of 10km and 15
  km have now also been successfully tested
• Some initial tests were conducted on a 5km length
  of fiber, transmitting a 20 GHz beatnote

9
Test Setup for measurement of phase drift between
the near and far end of the fiber. The fiber
length is actively corrected by means of an
optical interferometer servoed off of the
round-trip phase of a highly stable and coherent
master laser.
10
Phase Drift of 18.6 GHz signal going through 5
km of fiber. The fiber length correction is "ON"
for 660 seconds and then is turned "OFF".
11
Same test as previous figure, showing a closeup
of the measured residual phase with the
correction "ON"
12
Phase Drift of 18.6 GHz signal going through 5
km of fiber. The fiber length correction is "ON"
for 400 seconds and then is turned "OFF". This
plot also shows the correction voltage to the
fiber stretcher, which has a scale of 165 microns
per volt. (Right vertical scale is volts)
Figure 5 - Zoom of corrected portion of Figure 4.
13
(No Transcript)
14
10 km
15
(No Transcript)
16
Plot of Phase Drift of 18.6 GHz signal going
through 0 km of fiber. The fiber length
correction is "ON" for 800 seconds and then is
turned "OFF". The fiber stretcher voltage has a
scale of 165 microns per volt.
17

• The preliminary experiments have successfully
  demonstrated the transmission of a 18 GHz
  reference over 5 km with a residual phase
  fluctuation lower than 0.22 degrees RMS (33 fs
  RMS) over 10 sec. These short term fluctuations
  are at the limit of our measuring system.
• Further measurements are planned at higher
  frequency in order to further reduce the measured
  RMS phase residual.
• Over a longer period of time, 600 sec, the total
  peak-to-peak drift was less than 100 fsec.
  However, this was similar or only slightly larger
  than what was measured for a zero-length of
  fiber, so we conclude that part of the source of
  the drift is in the test setup.
• Additional drift may be due to the polarization
  or birefringent effect of the coarse fiber
  stretcher.

18

• This is the first demonstration of the ALMA line
  length correction technique to instrument-limited
  accuracy. The main remaining objectives are
• To prove the technique to greater accuracy
• To prove the technique with a moving fiber wrap
• To prove the technique in a field-system, such
  as the prototype antennas, esp. with buried fiber
  rather than a spool
• To improve the master laser coherence so that up
  to 18 km of fiber can be stabilized and better
  immunity to environmental perturbations results.
• Other subsystem issues
• Polarization maintaining components, or
  polarization alignment servo on the source end
• Minimization of polarization dispersion of the
  other compoents and the fiber, esp. fiber
  isolators required at the antenna end