J. Kissel

I've completed the weekly charge measurements; results attached below. I've also flipped the ETMX and ETMY bias signs, as per our plan for testing if ISC and CAL can handle the collateral damage of mitigating charge in this fashion. For now, I've only flipped the bias, and am waiting to see if what we coded in to guardian two weeks ago (see LHO aLOG 27890) will take care of the rest of the settings changes that account for that flip.

Just for clarities sake, since I *know* where gunna start losing track of which direction the bias was flipped -- today the biases were flipped as follows:
    ETMX from -9.5 [V] to +9.5 [V]
    ETMY from +9.5 [V] to -9.5 [V]

Though we had originally intended to flip the bias signs every week, last week's maintenance was far too hectic so Betsy was only able to get charge measurements at the last minute on Wednesday (see LHO aLOG 28093), and Thursday was too hectic with calibration measurements (and the IFO was already locked when I remembered) that we forgot to flip the bias sign. So, we flip this week on schedule.

As we have found when we'd done rapid flipping last summer, the charging rate after a sign flip is unpredictable. For example, though it's difficult to quantify with only two data points, between the last flip and today the charging rate is rather fast on ETMX (5 [V / week] as opposed to 2.5 [V / week] prior to the last flip). This is the sort of thing we need to keep an eye on, lest we again forget for a week, or the apparently-random-walking of charging rate starts, on average, bringing us to accumulate charge of a particular sign. This might also indicate that flipping the bias sign every two weeks (instead of every week) is better, given our frequency of charge measurements and the need to have several before establishing a trend.