Georgia, Robert

I have found minima in coupling of noise injections onto the building ground at EX (bias at +128) and EY (bias a +115) (https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=67075), and was recently able to investigate the corner station.  Grounding noise coupling at the corner station depends on two biases, ITMX and ITMY, so I first tried iteratively to find the minima, but the minima seemed to change. Further investigation suggested that the reason for the apparent variation was common mode suppression of the coupling, that is, that coupling was minimized whenever both test masses had about the same bias. Figure 1 shows that coupling is low when both are set either to +428 V or to -428 V but is high when one is at -428 V and the other is at +428 V.  The residual coupling is lower when both are set to -428 V than when they are both set to +428 V, so I would be inclined to leave them boh at the -428 V setting that Georgia had found produced lower noise. I didn’t get a chance to check the residual coupling when both were set to zero. 

The apparent common mode rejection suggests either that the effective charge is about the same for each of the ITMs, or that charging is not that important in the coupling.

Figure 1 also shows that the injected peaks at 11 and 22 Hz (I inject an 11 Hz comb) either do not respond to the bias setting (at 11 Hz) or have a minimum coupling level below which bias does not affect the coupling (22 Hz). This suggests that, at these low frequencies, coupling can be dominated by coupling higher up the test mass chain. I will have to investigate this low frequency coupling when I return. The end stations differ in that they do not appear to be dominated by coupling higher in the chain at the lowest frequencies. The amplitudes of  11 and 22 Hz peaks decrease with bias even at the lowest coupling levels. So for the ETMs the coupling can be minimized with bias even for low frequencies.

Figure 1 shows peaks produced by injecting a signal onto an ISC Common Slow Controls chassis, but I also found that the apparent common mode suppression and the insensitivity to bias at low frequencies is evident when I inject onto a cable tray (for both cases the return was from the beam tube), but at a lower coupling level.