Today, Sheila and I decided to increase the ETMX ESD bias voltage. We would like to operate at a higher bias for a short period to see if the extra actuation range allows us to survive the ETMX glitch locklosses. However, we also wanted to understand what effect this has on our range.

Sheila made small steps to increase the ETMX bias voltage and adjusted the L3 drivealign gain accordingly. After every step, I measured the DARM open loop gain and we adjusted the drivealign gain further to maintain the DARM UGF. Once we doubled the voltage, we further adjusted the drivealign gain to bring kappa TST to one. It's important to note that we saw that the correction factor required to maintain the UGF (at 70 Hz) was slightly different than the correction factor required to bring kappa TST back to 1 (measured at 17.6 Hz).

We took a PCAL broadband measurement at the regular and double bias configurations (documented here). Even with kappa TST near 1, there was a small frequency-dependent difference between the two measurements.

Sheila then took us back and forth from the double bias to regular bias states so we could get some noise comparison times to determine if the noise is worse with double bias.

I exported the PCAL broadband injections and used them to calibration GDS strain for the four times above into PCAL meters. The noise from 15-400 Hz looks to be the same for each time, so I don't think there will be a significant impact to the sensitivity if we double the bias.

I ran a linear regression fit on the voltage and drivealign data from our steps, using this document as reference. I was able to fit alpha - gamma (slope) and the beta values (y-intercept). I calculate alpha - gamma = 9.14e-10 N/V^2 and beta - beta2 = -4.08e-8 N/V

We are planning to go to this double bias for some time to see if reduces the number of ETMX glitch locklosses.