- Joe, Deepak, Volker, Cheryl

HAM2 IO install was completed last Friday.  The PRM surrogate was swapped to have the HR/HR optic, and to steer the beam to the parking position through the HAM2 top viewport.  MC3 OSEM protection baffle was installed using the template sent from LLO.  Minimization of the beam once the HWP after IM1 was installed reduced the beam to 7uW at PRM surrogate, which is a good thing since we're still sending the wrong polarization into the mode cleaner.

MC1 lower stage OSEMS were adjusted, so MC1's hanging position is unknown, so this morning I'll be working on it's alignment to get the mode cleaner flashing.

There is a black residue in cleaned chambers that comes off on latex gloves because they stretch further into the crevices that the wipes used to mop up the cleaning residue (here). We suggested in that alog that we test nitrile clean room gloves because they might not stretch into the crevices as much as the latex gloves. Jodi procured some nitrile gloves and when I was in HAM9, I compared the two types of clean room gloves. The photos show that much less residue came off on the nitrile (whiter) gloves than on latex gloves for the same pressure and length of rub against the HAM9 wall. I tested the 3 types of nitrile gloves (Hand PRO series 9100 and 7100 clean class nitrile,  and Valutek VTGNUTPFB95) that Jodi procured and they were all equally better than the latex gloves.

Robert S. Jodi F.

Summary: Fine particulate is visible in all nozzles of the H2 IMC tube, but not in the nozzles of the HAMs near it, suggesting that the particulate came with the beam tube. I measured dust counts during activity in the H2 IMC tube and they were low, about like cleaned chambers. I also looked for particulate that might have moved from the H1 IMC tube into HAM3 and did not find any clear examples. So I think the particulate is fairly stable and it is safe to pump the H1 IO chambers.

A fine particulate was noticed in the LLO IMC tube nozzles (here). Mike L. asked Jodi and I to go into H2 HAM9 and the new H2 IMC tube because there has been very little activity in there and we might be able to find if the particulate came from the IMC tube. The H2 input optics area has not been pumped down since the new tube was installed. Figure 1 contains photos comparing nozzles in the H2 IMC tube and HAM3. The top left photo is of particulate in the 4^th vertical nozzle from HAM9. I have drawn a face in the particulate with my finger. I found this particulate in every nozzle of the new MC tube that I examined. I inspected every nozzle of HAM9 and 8 and found none of this type (there were a few isolated  big pieces). Thus this particulate seems to have come in with the IMC tube.

To see how much particulate would be stirred up by activity, I did my scuffing walk test. The level in the MC tube averaged 140 0.5um particle counts per cubic foot, comparable to the cleaned chambers and much better than the ~1000 count levels from walking in the Y-Manifold (here).

Mike also asked me to go in and examine HAM3 to see if the vacuum cycling for leak checking had spread this fine particulate into the HAMs. The photo at the bottom of Figure 1 is a photo of a HAM3 nozzle, showing that there is very little dust compared to the H2 IMC Tube nozzle above it. Also, the few particles seem larger. I sampled the particulate at the HAM3-spool junction for elemental analysis to compare to the samples I took in the H2 IMC tube. I think it is safe to cycle the IO chambers because the particulate does not seem to have spread much – the burden of larger dust particles and especially silver metal particles near optics in HAM3 appears to be worse. 

Robert S., Jodi F.

The crystal chiller circuit has been increasing the motion of the PSL table by factors of 2 to 10 between 70 and 800 Hz (here). The vibration level varies in time and we have had some success by removing air from the manifold etc., but, until now, we have made no dramatic improvements.

In a recent alog (here) we showed that the quick-connect fittings in the high-flow crystal chiller circuit produced vibrations that were orders of magnitude larger than for barbed connector fittings, possibly because the nozzle in the quick-connects causes a small stream of water to spray chaotically and with high velocity inside the tube, and we suggested replacing the ones on the crystal box manifold.

Oliver Punken was here this week from UTB to work on PSL tasks, so we replaced quick-connects in 2 stages. In the first stage we replaced the connectors on the crystal box manifold, and straightened out the flow path (Figure 1). Figure 2 shows that the problem virtually went away after replacing these two connectors.  We also replaced the quick-connects and straightened out the flow in the manifold under the table (Figure 3) but made no further visible improvements (we were hoping to reduce the small residual at 100 and 150 Hz).

We may face future beam jitter problems from the chiller circuits, but the table vibration problem that we had identified is essentially solved. We recommend that LLO replace the two quick-connects on the manifold at the crystal box. They may also want to replace the two on the manifold under the table to keep both sites the same. 

Robert S., Michael R., Oliver P., Rick S.

Summary: No drop in magnetic coupling level over the 2 – 10 Hz measurement range was noted after removing eddy current damping magnets on M0 and L1. This was not unexpected because we had previously noted that there was also excess coupling at L2. The slit L2 parts are here and I suggest that we install these on ITMY and measure coupling again when possible.

We recently (here) found that large magnetic injections produced motions of ITMY that, when scaled linearly, suggested that ambient magnetic fields would produce motions that were orders of magnitude larger than our sensitivity goal at 10 Hz. Further tests (here) suggested that the coupling level was the same when SUS and ISI cables were disconnected and HEPI shut down, that the coupling was similar at ETMY, and that the coupling occurred at multiple suspension levels. We also found that coupling was within a couple of percent of linear for magnetic fields that were varied by a factor of 3 at both the ITM and the ETM.  Comparisons of BOSEM and OSEM injections to magnetic injections showed that coupling was excessive at L1 and L2 as well as M0, so the suggestion was made that in addition to removing the M0 ECDs, we also remove the blade spring ECDs and that we slit parts around the AOSEMS at L2, the penultimate level, in order to prevent eddy current circuits (here). For the test reported here, we removed the ECDs, but did not replace the L2 parts, so we expected coupling at L2 to be unchanged.

Figure 1 shows that coupling, as measured by the optical lever reflecting off of the ITMY test mass, has not significantly changed. For an ambient magnetic field background of 10 pT, the graph predicts about 1e-16 radians/sqrt(Hz) of motion. The point at 63 Hz measured in Aug. has been left off of the graph because there was no clear indication of the 63 Hz peak in the one 12 hour injection I had time for this month.  For the previous measurement at 63 Hz, cables were installed and damping was on. Thus one possibility is that the point at 63 Hz may have indicated coupling to the cables, but not directly to magnets on the quad. For all measurements here, except for one of the 3.5 Hz measurements, the SUS cables were disconnected and ISI and HEPI were off.

As a next step, I suggest that we install the slit L2 parts (Betsy says they can be installed in situ), and re-test magnetic coupling when possible.

Optical lever calibrations

We used two independent methods to re-calibrate the optical lever. First, the original DC method, calibrating beam spot motion by moving the quad diode on its translation stage and converting displacement to angle using the geometry. We also used an AC method to cross check the calibration; we injected 5000 counts at 1 Hz into M0 pitch or yaw, measured the beam spot displacement at the photodiode using a scale, and then calibrated the photodiode output using a 500 count injection (to be in the linear region of the photodiode) and assuming the amplitude of beam spot motion was 1/10 of the motion measured for the large injection (we confirmed actuation linearity with a 2500 count injection). The independent calibrations differed by 4% in yaw and 15% in pitch. We used the average of the two calibrations for the gain; Pit: 23.38, Yaw: 24.74. The August gain values are still on the MEDM screen because I did not want to change gains in the middle of collecting data, so the calibration was corrected off-line.

Peak disappeared when beam was blocked

We had previously tested for coupling of our injected field directly to optical lever electronics by blasting the electronics with large magnetic fields (here). This time I checked by blocking the optical lever beam while injecting a 5.5 Hz magnetic field. A flashlight was used to bring the signal level from the diode segments up to their nominal signal levels either in total RMS or at 5 Hz. The 5.5 Hz peak in the optical lever signal disappeared in both cases.

Set up

The setup was similar to what was used before. The attached photos show the coils. We ran 12 amps through these coils for most injections, producing magnetic fields at the test mass of several tens of micro Tesla. We have measured injected fields inside a BSC for a similar coil configuration and found the ratio of gradients to fields for the injection to be about the same as for ambient 60 Hz fields (here). The magnetometer, used in determining radians per Tesla of injection, was placed 10 cm under the center of BSC1. 10X attenuators were used and the amplifying filter box was bypassed in order to get the large magnetometer signal on scale. The magnetometer was calibrated in situ using a small calibration coil.

Robert S., Thomas V.

JimW, HughR, HugoP,

Finished payloading HAM2-ISI.

Set up the lockers. Lockers needed an horizontal set up only. We did not change shims. The Locked/Unlocked shift is comparable to HAM3-ISI's: way below requirements.

Requirements are (E1000309-v12): 
shift<1600cts 
Readouts<2000cts

The Horizontal CPS readouts are slightly out of requirements on 2 horizontal CPSs (H1 and H2). Targets are a bit too close to their CPSs, but they are far from contacting, so there is nothing to worry about. The zero of these horizontal CPSs will be reset once the payload, and thus the lockers, are in final state (after the IMC test).

Only the blade spring profiles of corner 1 and 2 could be measured. They are within requirements.

Requirements are (E1000309-v12): 
Flatness<15mils

We checked the level of the Optical Table with the optical Level: +-0.1mm.  More or less right at the 100urad requirement.

Locker shims and mass budget are recorded. The cables were dressed and HAM2-ISI was left unlocked, ready for testing on Monday.

The SUS-to-ISI WatchDogs IPC communication is currently bypassed with a constant in SUS's MC1 and MC3 models. The constant is 0, which correspomds to OK state of SUS. A screen-shot of these models is attached.

In this configuration, MC1 and MC3 cannot trip HAM2-ISI. In order to safely actuate on HAM2-ISI, this constant needs to be removed, and connections need to be made in the following models: 

• h1susmc1.mdl
• h1susmc3.mdl

Note: These constants were put in to avoid tripping HAM2-ISI during the chamber-side tesing of MC1 and MC3.

Attached are plots of dust counts > .3 microns and > .5 microns in particles per cubic foot from approximately 6 PM Dec. 13 to 6 PM Dec. 14. Also attached are plots of the modes to show when they were running/acquiring data.

Repairs OK, new joints OK -> still have a leaking 4.5" blank on BSC2 dome ((1)16.5" x (3)4.5" cluster) from initial leak testing that needs to be fixed.  This is the only joint on this volume (Vertex minus HAM1, 2 and 3) that remains to be fixed (that I am aware of - reminder, if you install a new flange, you need to label it as "NOT LEAK TESTED" otherwise I won't know it hasn't already been tested)

Again, for my own record, next week we need to:

Vacuum off suspensions for localized particulate.

Set lower stage AOSEMs to 50% OLV.

Remove FC and spot clean PR2.

Sopt clean MC2.

Swap PFA EQ stops with fluorel tips.

Set EQ stops and lock all nuts.

This afternoon we finished prepping MC1 and MC3 for the pump down scheduled for next week.  As per my prior alog list, we:

Tweeked the lower stage AOSEMs into place (Travis did the hard ones via YOGA moves across the table after I gave up).

Swapped the PFA EQ stops with the fluorel ones.

Installed the VAs.

Locked all the EQ stop nuts.

Add-on task - we dabbed a few particles off of the MC1/3 faces and then decided it was futile and will have to wait until the next vent for FC cleaning.  Note, while adjusting the AOSEM upside-down teensy set screws which are an inch away from the optic facees, we might have brushed a gloved knuckle against the MC1 AR optic.  Uugh.  Chalk that one up to an addition to a design change list or at least a lesson learnt.

Adding to the to-do/install list, Richard is also going to try to tackle the ground loop checks on HAM2/3 next week.

We'll also finish pump-prep work on the HAM3 SUSes (PR2 and MC2) next week.

• 6.3mag earthquake in Baja Calif at 2:36am
• Optical Levers:  (Robert, Thomas) Another calibration measurement (cover removed from viewport)
• Cheryl/IO checking ifo beam alignment on beam dump at HAM2
• Ski swapping prefilters for OSB area (including Optics Lab, Bake Oven Room---Cleanrooms/flow benches will be functioning, but will monitor dust monitors).  This took about 10min. 
• Surprise expressed at ITMy whose Sattelite Box was powered off last night (this was Robert last night for field tests).  Satelite box powered back up and ITMy damped.
• HAM2 SUS Watchdogs reconnected to SEI HAM2 &3.
• Michael & Oliver turned off the laser to do work on the water distribution system
• Ski may cause FMCS alarms at mid-stations

The Final In-Chamber Testing Report of HAM3-ISI was validated today. It is available under the DCC, as the 3rd section of E1200507 - HAM3-ISI Phase II Testing report (pages 35 to 58)

HAM3-ISI is now locked so the first contact could be peeled off.
 

JimW, HugoP,

Went in HAM2 to dress the cables that go from Stage-0 to Stage-1. The goal is to make sure they will not be rubbing.

HEPI will not be released before the next opening of the chamber. Hence, cables from stage 0 to the feedthrough were not dressed.

JimW, HugoP,

Locked HAM3-ISI so:
   - RobertS can take samples to evaluate contamination
   - SUS can remove first contact when ready

WP#3605, userapps svn r3464

I modified the h1iopseih23 model to reconnect the WD signal from SUS H2A. This permits MC1 and MC3 HAM2 OSEM signals to trip the DACKILL on SEIH23, along with PR2 and MC2 from HAM3. The MEDM screen was also changed accordingly.

All models on h1seih23 were stopped and restarted. ISI was burt restored to its safe.snaps (note, there seems to be problems with ISI HAM3's safe.snap). HPI models dont have safe.snaps and were restored to the backup made at the time of shutdown.

Yesterday, the Apollo finished wipe down, second vacuum, chamber FTIRs, and close-out inspection. Two doors were returned to the chamber, one torqued. This morning, the technical cleaning crew is doing first cleaning on the dome cleanroom. Once that is complete, the cleanroom will be moved over the dome and second cleaning will be done. The Apollo crew will move the dome spreader from the OSB to X-end and start torquing the remaining door bolts. After second cleaning, the dome will be cleaned. If time permits, we'll try for putting the dome on today. 

Closed the O-ring valve and the gate valve @ the turbo overnight (closed at 19:45 local time), and I will restart pumping tomorrow morning.

- Jodi, Cheryl

Last Thursday (Dec 6) Jodi and I inspected the H1 input beamtube.  Pictures attached show that we found white powdery stuff (HAM3 table, beam tube, MC2 cage), white fuzz (beamtube), part of a latex glove (baffle), fibers, and uniform particulate on the overhead nozzles (picture of gloved finger shows what came off with a swipe).