Following on from alog 42302, I have done an effective bias voltage measurement of ITMY.

The effective bias voltage appears significantly lower for ITMY compared to ITMX:

Note that the coupling from the signal drive to the optical lever, particularly in pitch, is reduced compared to the ITMX measurement. This is shown in the first attached figure, which shows the signal drive to optical lever transfer function as a function of bias voltage offset. This can be compared with the figure attached to alog 42302.

For these measurements, the ITMY optical lever was quieter than the ITMX. The final two attached screenshots show the ITMX and ITMY optical lever spectra respectively. Red traces (references 0 and 1) are without a signal drive, and blue traces are while driving the LL quadrant at 0.8 Hz, with a large negative bias voltage. The signal-to-noise ratio is much better on ITMY.

Greg, Dave:

following the failure of the E18-0 Pod-1 Disk-6 last week, Nexsan sent a replacement 6TB HDD today. Greg and I hot swapped the disks at 13:00 and added the new disk to the raid at 14:50. It is in the process of being rebuilt, which for a 6TB drive in background mode could take some time.

This was a truly hot swap, with the h1fw0 frame writer running throughout with no problems.

With the most current DRMI alignment, we found it hard to close the OMC ASC loops and had to put a few offsets in the AS WFS DC Centering loops in order to keep the OMC suspension from railing. 

Since the AS WFS weren't holding any particular reference, I took the time to offload the beam into the OMC using OM1 and OM2, of course this mis-centers the AS WFS so then I pico-ed the beam back onto AS_A and AS_B.  This leaves about 80% margin on the OMC_SUS DACs.  I then offloaded the control signals for the DC_Centering loops and the OMC_ASC to their respective alignment sliders.  

Also, I realized the the OMC guardian state "ASC_QPD_ON" doesn't reflect the OM actuation sign flip so I changed that as well and committed to SVN.

WP7619

Sheila, Jenne, Dave:

New code was installed for the models h1ascimc, h1susmc[1,2,3]. New code and new IPC senders were added to h1ascimc, associated IPC receivers were added to h1susmc[1,2,3]. The DAQ was restarted.

This is a late alog about what was done on Friday last week.

At first the IR beam was clipping on the right  side of the first 1" mirror downstream of the periscope (ALS-M3 on D1201103). I moved it by about 6mm toward the PSL.

I aligned the IR beam going into the SHG by using ALS-M3 and ALS-M5 so the green beam falls on  ALS-COMM diode.

The beam was centered on IR monitor diode using ALS-M4.

The centering on the green monitor diode was OK, there was no big change (if any at all) when I changed the angle of ALS-BS2.

Laser power upstream of ALS-BS1 was about 200mW, and the green power downstream of the IRHR dichroic was about 220uW. Before laser swap, IR power was about 1.2W according to IR_DC_POWERMON, and the green power was about 8.3mW. It seems like the numbers make sense, i.e. 8.3mW * (200mW/1.2W)^2 = 8.3mw/36 = 230uW ~ 220uW.

I don't know if we're throttling power using HWP and polarizing splitter right after the PMC.

I went to EX, turned on the ALSX laser and locked PLL.

EX is still in laser safe.

I had to manually set the slow feedback offset to about negative 800MHz until the beat note appeared on the correct side (lower than PSL). This was relieved by increasing the temperature setpoint by 0.26 C (23.67 to 23.93).

   Both filter are clean and clear. No debris or discoloration noted in either filter. 

   Additionally - Added 50ml water to the crystal chiller. Did not add any water to the diode chiller. 

   Closing FAMIS task #8309.   

On Bubba's request I have bypassed the RO cell phone alarms for 24 hours

Bypass will expire:
Tue Jun  5 12:39:02 PDT 2018
For channel(s):
    H0:FMC-CS_WS_RO_ALARM

14:06 UTC Vanessa to end X
15:26 - 15:35 UTC Jeff B. to/from cleaning area
15:47 UTC APS through gate
15:55 UTC APS through gate
16:49 - 17:33 UTC Jim to end X to pick up equipment
17:22 - 17:26 UTC Kyle to end Y VEA
17:35 - 18:37 UTC Karen to end Y to drop off supplies, then mid Y to clean
17:49 - 18:03 UTC Chandra to LVEA
18:13 UTC Georgia to LVEA to reroute cables for ITMY charge measurements
18:20 UTC RO Water alarm
18:28 UTC Keita to end X to turn on ALS laser (WP 7618)
18:51 UTC Richard to LVEA to run fiber

18:55 UTC Handing off to Corey

Preparing to pump end X
In process of opening GV12
Pumping end Y

DRMI/EY commissioning
DRMI locking, ASC WFS not converging
EOM swap -> possible phase difference

PSL ISS commissioning

CP ion pumps installed and pumping

ISCT1 table alignment ongoing

mid Y CP4 oven work is done

Chiller in fiber welding lab died, moving one from end X

Tuesday: LASER SAFE
Unlock HAM 5/6 HEPI, check range of motion
Pulling fibers in MSR (WP 7617)
Swapping controllers for illuminators
OMC/AS-C whitening chassis swap
Resume filter swapping of end X HEPI

I've made a similar measurement to the usual ETM-optical-lever-effective-bias-voltage, on ITMX.

Method

Since the ESD quadrants are hooked up to a low voltage driver, in order to drive them hard enough I connected the bias output of the ITMY HV chassis to the quadrants directly, one by one. I looked at the optical lever response to the individual quadrant drive (V_{signal}) as a function of ITMX bias voltage. The bias voltage where the optical lever response is zero is the usual V_eff we measure.

I drove each quadrant for 2 mins, at 0.8 Hz, with an amplitude of ~120000 counts as monitored on H1:SUS-ITMY_L3_MASTER_OUT_DC_DQ. I recorded the timestamps for these drives and bias voltages and then looked back at the optical lever pitch and yaw channels. I then calculate the transfer function from the signal drive to the optical lever, and plot this as a function of bias voltage. See attached plot.

The scripts (driveITMChargeMeas.m, using the function process_single_quadrant.m) live in userapps/release/sus/common/scripts/quad/opLevChargeMeasurements/ and are adapted from Sheila's in-lock charge measurements.

Final numbers

Note that due to the larger gap between the test mass and reaction mass, the alpha and gamma co-efficients are smaller for the ITMs than the ETMs (see alog 38387), giving a larger V_eff. To compare the V_eff of ITMX to the ETMs we need to divide by a factor of ~1.5.

I'm not 100% confident in this data as the V_eff varies significantly for pitch and yaw of the same quadrant (in particular UR) which I don't understand.

[Niko, Suresh]

Motivation:   

    We suspect that the He-Ne Laser power supply may be generating an unacceptable amount of EMI.  To check this we installed a magnetometer close to the power supply of the laser.

Updated He-Ne laser setup:

   1) we moved the power supply close to the laser (see pictures attached) and placed the magnetometer near them on the ground.

   2) the optical fiber was rerouted to minimise dangling.  The fiber has been supported close to the laser and the oplev transmitter to decrease fiber motion.

   3) To check if the large (39dB) whitening gain was introducing noise into the oplev signals.  We temporarily reduced the gain to 19dB and then after about an hour put it back to 39dB.  This change did not seem to have an effect on the noise level in the oplev signals. 

   4) The Magnetometer signals are connected (X,Y,Z) to inputs (1,5,6) on PEM BNC breakout panel which show-up in CDS channels H1:PEM-EY_ADC_0_08_OUT_DQ, H1:PEM-EY_ADC_0_12_OUT_DQ and H1:PEM-EY_ADC_0_13_OUT_DQ.

Results Pending further analysis:

Will look at data with Robert and report on the EMI issue next week.

Will look at oplev data to see if rerouting the fiber has made any difference.

Dan Brown, TVo

We relieved the AS DC Centering loops by changing the OM1 and OM2 alignment offsets so that when we lose lock on DRMI we'll be a bit closer to a decent alignment on AS_A which will help speed up re-locking.

Sheila, Gabriele, TVo

We found that the phasing wasn't very good for REFL 9 and 45 when locked on DRMI so we changed the tuning:

Decoupled vent/purge line and shut down purge-air supply.  I then opened the 2 1/2" vent/purge valve and let the ~1 psig air "blow down" to slightly above room pressure.  The measured dew point of the exiting air was -10C. 

After only 3 iterations of briefly driving GV12 in the OPEN stroke (1000 rpm - stop, wait a day, 500 rpm - stop, wait 3 hours then 400 rpm) 3 hours of accumulated gas from the isolated gate annulus volume was dumped into the Y2 BT module and CP4 (see attached).  This indicates that a "crescent" of O-ring has obviously pulled away but it is likely that much of the total circumference of the O-rings still remain adhered.  As such, we will continue our ultra-conservative approach and operate the motor for only brief periods so as to keep a slight tension on the O-rings that will encourage them to release on their own.   This, being in contrast to opening the valve in one step which could result in the O-rings being pulled from the capturing dove-tail grooves. 

It looks like the ion pump can handle the remaining annulus volumes unassisted by the turbo but I am leaving the turbo+aux. cart running (but isolated) over the weekend. 

[Mark, Tyler, Chandra]

Mark and Tyler installed IP3 this morning. It is leak tight (with 7.2e-10 mbar-L/s background). It is currently being pumped with aux cart + hung turbo. It can't maintain itself yet. Voltage gets to 2000 V (84 mA) before tripping off. Will valve it out from aux cart over weekend and resume pumping on Monday.

IP2 is slowly recovering (re-installed this old pump that had vented with room air). It gets up to 4400 V before tripping. Will also valve out over weekend.

While inspecting the HR surface during (and just) after the FirstContact pull on the ETMX HR Surface yesterday, Travis and I saw a few points within the central 6" of the optic.  Pictures are attached.  One feature looks more like a small sleek or scratch rather than a dot.  Upon review of the "map" we made roughly a year ago when the optic was in the bonding lab (also attached), we see that many of these were there during that inspection.  In conference with GariLynn this morning, she confirmed my decision to move to chamber closeout and the door has just been installed on the chamber.  We will digest these pictures and compare to previously obtained coating maps and metrology data.

Note, keep in mind that it is very difficult to say much about the optic surface quality based on these photos.  It is difficult to state whether such observed features will be a problem, whether they are in the coating or on top of the coating, etc.  TBC...

This afternoon,

- I wiped the AERM and ETM barrel surfaces as best I could with swabs and methanol to remove particulate.

- Travis and I removed the FirstContact sheet from the optic - upon inspection, we still found some point features within the central ~6" of the optic HR surface.  They did not blow away with 20psi N2 Top Gun, nor do they look like FC pieces.  I will compare my pictures to the map we made during last year's lab inspection.

- Travis and I checked for charge via the "normal" procedure of looking with an electrometer at the back surface of the AERM with a special bracket.  Details below to follow.

- Travis ocked all EQ stops nuts, taking into account sag once under vacuum.

- We set the top 4 QUAD BOSEMs to slightly more closed, also to take into account the sag of the suspensions once under vacuum.

- Removed all tooling, wiped floor and ACB surface.

- Travis swung the ACB back into it's nominal position.

- Jim unlocked the ISI and made a quick medm check.

- Laid new witness CC wafers and optics. - Pulled FC from 2x optics (1 vertical on QUAD, 1 horizontal under QUAD).

Tomorrow we will run TFs in the am and then launch the door crew.

files saved in:

• .../QUAD/H1/ITMX/SAGM0/Data
• .../QUAD/H1/ITMY/SAGM0/Data
• all TFs looked acceptable, with some noise, but nothing missing or added