Stefan Hild and A.Freise Advanced Virgo meeting, December 2008 Preliminary Thoughts on the optimal Arm Cavity Finesse of Advanced Virgo

S. HildAdvanced Virgo, 18th of December 2008 Slide 2 Overview  Advanced Virgo design: Many decissions have to been taken in near future.  One decision with high impact on other subsystems is the choice of the Advanced Virgo arm cavity finesse.  This talk is a preliminary collection of aspects that need to be taken into account. Intention: to discuss if our considerations are okay with everyone…

S. HildAdvanced Virgo, 18th of December 2008 Slide 3 Context  Advanced LIGO initially planned to go for an arm cavity finesse of ~1200.  The advanced Virgo reference design features an arm cavity finesse of 888.  Recently the Advanced LIGO team decided to lower the arm cavity finesse to 450.  One of the OSD tasks is to find the optimal arm cavity finesse for advanced Virgo.  Need to evaluate the benefit from lowering the finesse.

S. HildAdvanced Virgo, 18th of December 2008 Slide 4 How to compare different arm cavity finesse values?  A change of arm cavity finesse goes hand in hand with a change of the optical power inside the arm cavities.  If we decrease the arm cavity finesse, the stored optical power will go done as well. => stronger shot noise contribution. => not a fair comparison.  One can compensate for the lower finesse by increasing the power recycling gain.  Proposal: If we change the arm cavity finesse we will always restore the intra cavity power by increasing the power recycling gain, thus we always compare configurations with ~750kW per arm.

S. HildAdvanced Virgo, 18th of December 2008 Slide 5 Potential reasons for lowering the finesse?  Sensitivity ???  Mirror losses ???  Dark fringe offset ???  Noise couplings from small Michelson ???  Thermal load of BS, ITM and CPs?  Lock acquisition ???  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 6 Michelson sensitivity versus arm cavity finesse  In the initial detectors the arm cavity finesse determines the detector bandwidth:  Low finesse = large bandwidth  High finesse = best peak sensitivity  Is this also true for Advanced Virgo?

S. HildAdvanced Virgo, 18th of December 2008 Slide 7 Sensitivity for finesse 888 and 444  Let’s see how the ADV sensitivity changes if we lower the arm cavity finesse by a factor of 2. Step 1: double ITM transmission double PR factor Step 2: If we half the arm cavity finesse we also have to compensate the Signal Recycling parameters: double Signal Recycling detuning double SRM transmittance

S. HildAdvanced Virgo, 18th of December 2008 Slide 8  The Advanced Virgo sensitivity is (within a certain) range independent of the arm cavity finesse !! Sensitivity for finesse 888 and 444

S. HildAdvanced Virgo, 18th of December 2008 Slide 9 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ???  Dark fringe offset ???  Noise couplings from small Michelson ???  Thermal load of BS, ITM and CPs?  Lock acquisition ???  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 10 Finesse and mirror losses  Advanced Virgo preliminary design assumes 37.5ppm loss per surface.  This is an ambitious goal. What happens if the losses turn out to be twice as much (75ppm)? Any influence of arm cavity finesse?  The sensitivity changes with the mirror losses independent of the arm cavity finesse.

S. HildAdvanced Virgo, 18th of December 2008 Slide 11 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ???  Noise couplings from small Michelson ???  Thermal load of BS, ITM and CPs?  Lock acquisition ???  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 12 Dark fringe offset and arm cavity finesse  Consider imbalanced losses in the two arm cavities. => excess carrier light at the dark port.  Optical readout quadrature is determined by a combination of dark fringe offset and carrier light from differential losses.  To get best sensitivity we probably want to dominate the local oscillator by the dark fringe offset (need to check this…).  If strong light contribution from differential losses => need larger dark fringe offset => need to detect more light power.  Carrier light from differential losses at dark port proportional to arm cavity finesse.  In principle low finesse reduces the dark port power due to differential losses in the arm cavity. However, the dark port power from differential losses is rather small (<5mW) …need to put exact numbers on this..

S. HildAdvanced Virgo, 18th of December 2008 Slide 13 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ……………………………………………...YES  Noise couplings from small Michelson ???  Thermal load of BS, ITM and CPs?  Lock acquisition ???  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 14 Noise coupling from the small Michelson  All differential arm length noise inside the small Michel- son (MICH) gets suppressed by the arm cavity finesse.  Lower finesse => stricter requirements for:  Thermo refractive noise inside ITMs, CPs, BS.  Quietness of wedged optics (CPs? ITMs? BS?)  … etc …

S. HildAdvanced Virgo, 18th of December 2008 Slide 15 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ……………………………………………...YES  Noise couplings from small Michelson ……………...NO  Thermal load of BS, ITM and CPs?  Lock acquisition ???  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 16 Thermal load of BS, CP and ITM substrates  Optical power inside the power recycling cavity is proportional to inverse of the arm cavity finesse.  Lowering the arm cavity finesse from 888 to 444 increases optical power in BS, CP and ITM substrates from 1.35kW to 2.7kW.  Any problem from thermal lensing? …need to check with TCS …  ITM probably okay, since dominated by coating absorbtion. (?)  How about thermal lenses in BS and CPs ?

S. HildAdvanced Virgo, 18th of December 2008 Slide 17 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ……………………………………………...YES  Noise couplings from small Michelson ……………...NO  Thermal load of BS, ITM and CPs …………..………...NO  Lock acquisition ???  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 18 Lock-acquistion and finesse  The capture range of arm cavities inverse proportional to the Finesse.  Lowering the arm cavity finesse would make lock acquisition easier.  … need some quantitative input from ISC …

S. HildAdvanced Virgo, 18th of December 2008 Slide 19 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ……………………………………………...YES  Noise couplings from small Michelson ……………...NO  Thermal load of BS, ITM and CPs …………..………...NO  Lock acquisition ………………………………………………...YES  Losses inside the recycling cavities ???  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 20 Losses inside PRC and SRC  If there are unexpectedly high losses inside the PRC, then a high arm cavity finesse would be better.  If there are unexpectedly high losses inside the SRC, then a low arm cavity finesse would be better.  Advanced LIGO argumentation (from LIGO-T D):  PRC losses can be compensated for by higher laser power.  For equal PRC and SRC losses ALIGO favors a small finesse.  However, only ‘really’ strong losses (0.x%) inside the SRC seem to have a significant impact. … need to put exact numbers on this for Advanced Virgo….

S. HildAdvanced Virgo, 18th of December 2008 Slide 21 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ……………………………………………...YES  Noise couplings from small Michelson ……………...NO  Thermal load of BS, ITM and CPs …………..………...NO  Lock acquisition ………………………………………………...YES  Losses inside the recycling cavities ………….…not clear  Coating Brownian from ITMs ???  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 22 Coating Brownian and finesse  Lower finesse => higher transmittance of the ITM HR coating.  Lowering arm cavity finesse from 888 to 444:  increasing ITM transmittance from to  might be able to get rid of 1 or 2 coating layer on ITM  Reduce coating Brownian of ITM  Might change overall coating noise contribution by a few %. … need to check exact numbers … Coating Brownian noise of one mirror:

S. HildAdvanced Virgo, 18th of December 2008 Slide 23 Potential reasons for lowering the finesse?  Sensitivity …………………………………………………..independent  Mirror losses ……………………………………………...independent  Dark fringe offset ……………………………………………...YES  Noise couplings from small Michelson ……………...NO  Thermal load of BS, ITM and CPs …………..………...NO  Lock acquisition ………………………………………………...YES  Losses inside the recycling cavities ………….…not clear  Coating Brownian from ITMs ………….………………...YES  … anything else ???

S. HildAdvanced Virgo, 18th of December 2008 Slide 24 Summary  For the choice of the optimal arm cavity finesse several aspects need to be taken into account:  PRO low finesse: Lock acquisition, coupling of differential losses to dark port power, ITM coating Brownian, (losses inside the SRC?).  CONTRA low finesse: Noise couplings from MICH to DARM, thermal load of BS, CP and ITM substrate.  Within a certain range of values (finesse between 300 and 1000) none of the aspects seem give a NO-GO argument.  Main Question: Are there any strong arguments from any other subsystems for choosing a particular arm cavity finesse.