jeffrey.kissel@LIGO.ORG - posted 13:28, Wednesday 30 August 2017 - last comment - 14:55, Wednesday 30 August 2017(38457)
Third and Final Cycle of ETMX TMDS Complete
J. Kissel, N. Mavalvala, K. Ryan, D. Sigg We've successfully completed the third and final cycle of the Test Mass Discharge System on H1 SUS ETMY (BSC9/BSC5). All methodology and numbers ran very similar to the second session (LHO aLOG 38453), and the latter half of the first session (LHO aLOG 38430). Because I was operating the knobs, my electrometer square wave voltage when measuring the ion "current" was 22.4 V (the maximum), instead of the previous runs in which Nergis measured with 12 V. After noticing the difference, we checked the electrometer's measured ion "current" at both 22.4 and 12 V, and measured similar values, so I stuck with 22.4 (cause it was easier just to crank the dial to max vs. find where 12 V was on the scope). See the attached notes for any further detail. Also of note, we noticed about a factor of 2 (maybe 1.5) increase in electrometer-measured ion "current" (voltage) from +/-4.5 to +/-9 [Vp] when the TMDS system's scroll pump is ON vs. OFF (while the whole system is still valved off to the chamber). Maybe the scroll pump is sucking up ions from the electrometer? Still no explanation as to why we're getting a factor of 1.5 less "current" than LLO -- again, Ryan's report during their Jul 2016 TMDS run (LLO aLOG 35636, they were able to get upwards of +/- 12 to 13 [V]. Nergis suggests it's a function of calibration on the electrometer.
Non-image files attached to this report
Comments related to this report
The input impedance into the D1500061 electrometer board is 1e9 ohms. While this is relatively low in terms of the issues with contamination associated with electrometers having 10 to 100 times this input impedance, it is still feasible that somewhere in the chain of components from the pickup electrode to the input pin of the electrometer IC, that there is some parallel resistance associated with either moisture, or contamination. This would cause an apparent decrease in the measured current. A simple calibration might be possible by ordering a glass encapsulated 1e9 ohm resistor, cleaning it well, and using it as part of a voltage divider. The voltage divider formed by this glass resistor and the input impedance to the electrometer might be used to check the calibration via the normal electrometer output.