Displaying report 1-1 of 1.
Reports until 13:03, Tuesday 11 February 2020
H1 CDS (ISC, SUS)
filiberto.clara@LIGO.ORG - posted 13:03, Tuesday 11 February 2020 - last comment - 15:21, Friday 31 July 2020(55041)
Baffle Photodiode Amplifier Powered Down At EX

The following modifications were made at EX to help with the ongoing noise hunting investigation.

1. Baffle Photodiode Amplifier moved from SUS-R1 to FAC-R1
2. Baffle Photodiode Amplifier powered down
3. Beckhoff hub isolated from rack
4. ESD current limit resistor box removed and SHV barrels installed

The removal of the current limit resistor should have no impact. The installed ETM Low Noise ESD Driver Chassis (D1500129) has built-in current limit protection.
The Baffle Photodiode Amplifier unit was powered down to see if the Beckhoff terminal might be a noise source. Unit at EX is normally not used to lock the Interferometer. Unit can be powered back on by reconnecting the ±24V power cable.

F. Clara, R. Mccarthy, R. Schofield

Comments related to this report
jeffrey.kissel@LIGO.ORG - 13:58, Tuesday 11 February 2020 (55043)CAL, DetChar
Link 

Tagging DetChar and CAL.

These changes *should* not have any *bad* effect on the detector, but
    (DETCHAR) These changes were made in order to reduce the amount of spectral features in h(t) based on Evan's studies for the CW group (e.g. LHO aLOG 53439, G1902208) and the differences between H1 and L1.
     We would appreciate any help that @DetChar can provide over the coming days/weeks in assessing whether this change made a difference (whether that be positive, negative, or even "can't tell.")

    (CAL) The change in the resistor box -- "ESD current limit resistor box removed and SHV barrels installed" -- should, nominally, NOT affect the actuation strength of the ETMX ESD, but I'll be taking a look to see if the online tracking system *of* that strength made a difference.
jeffrey.kissel@LIGO.ORG - 16:16, Tuesday 11 February 2020 (55054)CAL, DetChar
Link 

The first observation ready segment after these changes were made started on Feb 11 2020 22:25:54 UTC, at GPS time 1265495172, or 14:25:54 PST.

Upon initial investigation, the DARM actuator's time-dependent correction factors (i.e. the point-estimate of the actuation strength relative to the 2020-01-03 model installed on 2020-01-13) are the same as prior to today's maintenance activity and thus I'm 90% confident that there has been no change to the actuator as a result of the electronics activity at EX today. 

Nice!
Images attached to this comment
jeffrey.kissel@LIGO.ORG - 15:21, Friday 31 July 2020 (56346)
Link 

R. Abbott, J. Kissel

Upon further investigation of calibration measurements taken of the TST stage over the remainder of the run, there appears to be a clear bifurcation in the data between the 2020-02-10 and 2020-02-24 measurements (see attached data collection). 

The bifurcation in the data is consistent with a ~4-5 deg change at 1 kHz. 

After consulting with Rich, we agree that -- given the residual parasitic cable capacitance between the chamber feedthrough and the ESD pattern itself (we guesstimate at ~600 pF), the act of replacing the former 10k current limiting resistors on the quadrant "signal" paths with the SHV connections (which have negligible resistance) removed the voltage-divider-like, passive low-pass response of a pole at 1/(2*pi*10e3*600e-12) = 26525 kHz, which *was* incurring a phase loss at 1 kHz of 180/pi*atan(-2pi * 1000 Hz * 10e3 Ohm * 600e-12 Farad) = -2.15 deg. Given the uncertainty in the cable capacitance, we could easily imagine a phase change of 5 deg after switching to ~0 Ohm SHV barrels.

The consequence: we'll just have to split up this collection TST measurements in O3B in to two periods, and run the standard fit for unknown systematic error separately.
Non-image files attached to this comment
Displaying report 1-1 of 1.