40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log  Not logged in ELOG logo
Entry  Mon Feb 11 19:53:59 2019, gautam, Summary, Loss Measurement, Loss measurement setup DQcheck.pdfinitialData.pdf
    Reply  Tue Feb 12 18:00:32 2019, gautam, Summary, Loss Measurement, Loss measurement setup 
       Reply  Tue Feb 12 22:59:17 2019, gautam, Summary, Loss Measurement, Y arm loss segmented.pdfconsolidated.pdf
          Reply  Wed Feb 13 02:28:58 2019, gautam, Summary, Loss Measurement, Y arm loss systUnc.pdf
             Reply  Thu Feb 14 21:29:24 2019, gautam, Summary, Loss Measurement, Inferred Y arm loss  modelPerturb.pdfposterior_modelParams.pdfposterior_Loss.pdf
                Reply  Sun Feb 17 17:35:04 2019, gautam, Summary, Loss Measurement, Inferred X arm loss  DQcheck_XARM.pdfconsolidated.pdfposterior_modelParams_XARM.pdfposterior_Loss_XARM.pdf
Message ID: 14448     Entry time: Mon Feb 11 19:53:59 2019     Reply to this: 14449
Author: gautam 
Type: Summary 
Category: Loss Measurement 
Subject: Loss measurement setup 

To measure the Y-arm loss, I decided to start with the classic reflectivity method. To prepare for this measurement, I did the following:

  1. Placed a PDA 520 in the AS beam path on the AS table.
  2. Centered AS beam on above PDA 520.
  3. Monitored signal from PDA520 and the MC transmission simultaneously in the single-bounce from ITMY config (i.e. all other optics were misaligned). Convinced myself that variations in the two signals were correlated, thus ruling out in this rough test any interference from ghost beams from ITMX / PRM etc.
  4. For the DAQ, I decided to use the two ALS Y arm channels in 1Y4, mainly because we have some whitening electronics available there - the OMC model would've been ideal but we don't have free whitening channels available there. So I ran long BNCs to the rack, labelled them.
  5. It'd be nice to have these signals logged to frames, so I added DQ-channels for the IN1 points of the BEATY_FINE filters, recording at 2048 Hz for now. Of course this necessitated restart of the c1lsc model, which caused all the vertex FEs to crash, but the reboot script brought everything back smoothly.
  6. Not sure what to make of the shape of the spectrum of the AS photodiode, see Attachment #1 - looks like some kind of scattering shelf but I checked the centering on the PD itself, looks good. In any case, with the whitening gains I'm using, seems like both channels are measuring above ADC noise.
  7. Found that the existing misalignment to the ETMY does not eliminate signatures of cavity flash in the AS photodiode. So I increased the amount of misalignment until I saw no evidence of flashes in the reflected photodiode.
  8. Johannes' old scripts didn't work out of the box - so I massaged it into a form that works.
  9. Re-centered Oplevs to try and keep them as well centered in the linear range as possible, maybe the DC position info from the Oplevs is useful in the analysis.

I'm running a measurement tonight, starting now (~1130PM), should be done in ~1hour, may need to do more data-quality improvements to get a realistic loss number, but I figured I'd give this a whirl.

I'm rather pleased with an initial look at the first align/misalign cycle, at least there is discernable contrast between the two states - Attachment #2. The data is normalized by MC transmission, and then sig.decimated by x512 (8**3).

Attachment 1: DQcheck.pdf  82 kB  Uploaded Mon Feb 11 23:30:50 2019  | Hide | Hide all
Attachment 2: initialData.pdf  18 kB  Uploaded Mon Feb 11 23:38:33 2019  | Hide | Hide all
ELOG V3.1.3-