Sheila, Georiga, Craig with help from Chris Wipf for the NB code and Erik Von Reiss for a tricky dtt issue.
On Friday we re-ran noise budget injections with Jenne's latest cleaning in, and the changes to the noise budget including calibration of CAL-DELTAL as described in 66304.
There are a few caveats to this noise budget:
- The shot noise isn't exactly correct because we normally apply the measured cavity pole and optical gain scale factor kappa_c calculated from the pcal lines. These aren't correctly calculated right now because applied a fudge to the calibration 66471
- There is no squeezing injected at this time.
- Previous versions of this noise budget have ignored the high frequency response of the DCPD whitening chassis. with the new whitening chassis this is a larger effect and needs to be corrected for. Chris helped me implement something that can take this into account for the DARM measured trace but we still need to do a correction for the dark noise and the laser noise projections.
- The dark noise hasn't been remeasured since the DCPD whitening chassis change.
- As usual the frequency noise injection is done by transitioning to REFL B and using REFL A as an out of loop sensor, so at low frequencies this will be an overestimate of the frequency noise by sqrt(2).
Things that are notable in these plots:
- SRCL noise is costing us some range.
- Input jitter is close to DARM from 100-200Hz according to the IMC PZT injection, this same noise isn't seen in the intensity noise projection suggesting that this isn't the usual coupling through misalignment of the ISS second loop array.
- Elenna and I had a look at the ASC subbudget in some detail, one of the main observations is that both soft yaw loops are contributing more than seems necessary.
- The projection says that laser frequency noise is close to DARM at 30Hz. We notch the frequecy noise injection at the OMC dither frequency to prevent down conversion through the OMC LSC, but it would be good to double check the linearity of this projection anyway.
For the total noise budget and the ASC budget, it is helpful to have the interactive svg plots that Craig made so you can turn on and off traces (to tell which is which). I've exported them here.
Attached is a plot that shows some rough projections of what our range would be with selected improvements. The 4.5 dB shot noise improvement is just that, a reduction in shot noise without degredation of quantum radiation pressure noise. We will work on including the filter cavity in a more realistic way, but that probably wouldn't make much of a difference to this projection which estiamtes we could be at a little bit better than 140 Mpc with our best squeezing and current IFO noise. There are also traces that show some other improvements, reducing LSC + Laser noise both by half would give us about 5 Mpc without squeezing, more like 7 Mpc with sqz. Reducing the residual (all unknown noise) by half would now give us about 8 Mpc. With all of these improvements together (including reducing unknown noise) we can get close to 160 Mpc.
