AdVirgo LSC requirements on accuracy and seismic isolation

Document Sample
AdVirgo LSC requirements on accuracy and seismic isolation Powered By Docstoc
					AdVirgo LSC: requirements on
accuracy and seismic isolation
              Gabriele Vajente
              AdVirgo (bi)-weekly meeting
              October 16th 2008
                 Summary


         Requirements on residual RMS motion of longitudinal
         degree of freedom
              Fringe width
              Simple estimate of up-conversion
              Actuation range
         Requirements on seismic isolation of core optics


         VIR-XXXX-08, “Requirements for the Advanced Virgo
         Length Sensing and Control system”



Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   2
                 Fringe width


         Residual RMS to be 100 times below cavities line-widths
              DARM < 3 x 10-12 m
              PRCL < 9 x 10-11 m
              SREC < 2 x 10-10 m
         MICH has a similar effect to DARM, given a de-
         amplification by the cavity finesse 2F/pi = 560
              MICH < 1.1 x 10-11 m
         CARM has a similar effect to PRCL, given the cavity
         amplification
              CARM < 1.3 x 10-14 m OR 1.3 mHz




Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   3
                 Static offsets


         DARM has a static offset for DC read-out of 10-11 m.
         RMS should be a factor 100 below this offset
              DARM < 10-13 m
         SREC has also an offset of about 0.1 rad = 3.4 x 10-8 m
              SREC < 3 x 10-10 m




Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   4
                 Noise up-conversion / DARM

        DC read-out introduces an
        intrinsic non linearity of ASY_DC
        signal w.r.t. DARM motion:




        This creates a quadratic term
        which can up-convert the low
        frequency RMS around large high
        frequency lines


Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   5
                 Noise up-conversion / DARM


         Require a line 100 time above noise floor to create an
         up-converted noise a factor 10 below the noise floor




                                                                         100
         DARM < 2 x          10-14    m

                                                                           10




                                                      0             Frequency

Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008               6
                 Noise up-conversion / Aux DOFs
         Residual motion of auxiliary dofs can change power on
         dark fringe.
         Modulation of ASY_DC optical gain.
         Another possible mechanism of up-conversion:



         MICH / PRCL are tuned at resonance, quadratic
         coupling of RMS motion
         SREC is off-resonance, linear coupling

         MICH < 9 x 10-11 m
         PRCL < 1.5 x 10-10 m
         SREC < 1.4 x 10-9 m

Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   7
                 Summary of residual RMS requirements




        Note: AdLIGO documents gave lower limits, motivating them with photo-
        diode saturations. We never saw such a thing in Virgo, but we should
        investigate it for AdVirgo



Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008               8
                 Actuation range
         Residual motion of auxiliary dofs create a signal on
         ASY_DC because of off-diagonal terms in optical matrix
         These are external perturbations at LF, completely
         corrected by DARM loop
              These signal are directly translated in DARM correction
         At low frequency:


         To avoid signals from aux degrees of freedom to
         increase DARM correction, one requires these false
         DARM motion to be below the real one (0.1 – 0.2 mm
         RMS)




Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008       9
                 Effect of noise subtraction paths


         Cfr. VIR-050A-08 “Longitudinal noise subtraction: the
         alpha-, beta- and gamma-technique”
         Optimal noise subtraction must be used in AdV to reduce
         the coupling of sensing noise
         The side effect if the increase of any noise at the
         actuation level by a factor which is equal to the auxiliary
         loop closed loop gain
              Having an optimally tuned alpha/beta/gamma/delta technique is
              equivalent to coupling auxiliary degree of freedom seismic noise
              as if there was no auxiliary loop closed loop gain




Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008                10
                 Actuation range / 2




         For CARM this is a requirement on the CARM closed loop residual
         motion
         For MICH, PRCL, SREC it is a requirement on the open-loop low
         frequency RMS motion
         Demanding for SREC
         It may be relaxed implementing a High-pass filter for noise
         subtraction path
              To have a subtraction ~100 at 10 Hz, the HP filter must induce a phase
              rotation at 10 Hz smaller than 0.5 degrees (NOT EASY)


Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008                      11
                 Seismic motion requirements


         Seismic motion at the level of auxiliary degrees of
         freedom would be suppressed as much as one wants by
         the auxiliary loop, at the price of re-injecting more
         sensing noise
         But the use of optimally tuned noise subtraction paths re-
         introduce auxiliary dofs seismic noise
         The seismic motion of MICH/PRCL/SREC is
         reintroduced in the gravitational channel with the
         couplings given by the optical matrix and no suppression
         due to the auxiliary loop effect



Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008     12
                 Seismic noise coupling to h




Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   13
                 Requirement on seismic noise

          Given by (design h) / (Optical TF) – no safety factor




Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   14
                 Seismic requirements
       Assuming a safety
       factor of 10, the
       requirements at 10
       Hz are:


      10-12 m/rHz




                             (Dominique Huet)


Gabriele Vajente, AdVirgo (bi)-weekly meeting – October 16th 2008   15

				
DOCUMENT INFO