jason.oberling@LIGO.ORG - posted 09:26, Tuesday 30 May 2017 (36513)
PSL Weekly FAMIS Tasks (FAMIS 3652 & 8424)
TITLE: 05/30 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Maintenance
OUTGOING OPERATOR: N/A
CURRENT ENVIRONMENT:
QUICK SUMMARY: ops lazy script isn't doing transition (-t).
After Sheila left today, I tried locking the interferometer several times with no success. The below are some notes.
The ice blockage has opened up on CP3! See last VE log entry. Kyle and I connected full T-sized GN2 bottle with acrylic rotameter (5 LPM range) to the sensing line and started out flowing 1 LPM. The flow held and we noticed fluctuations in the exhaust pressure (0.2-0.4 psi). Then I got antsy and bumped flow up to 1.5 LMP and it slowly increased to 2 LPM. The signal value (% full) is slowly growing! We saw exhaust pressures as high as 0.8 psi and signal value up to 10%. We are leaving it flow at 2 LPM overnight with shift-monitoring and -logging. Kyle will log odd hours and I will log even hours. We are being overly cautious about the not boiling CP3 dry overnight.
0900 hrs. local 8.9% full @ 0.8 psi exhaust. PT243 1.8 x 10-9 torr This is my last update for this aLOG entry - on my way to LHO
2302 hrs. local CP3's level indication is 4.7% full with 0.5 psi exhaust and no change to PT243.
Midnight: 3.6% and 0.4 psi exhaust lressure. No change on PT-243 (2.0e-9 Torr). Exhaust temps are both 21C.
5/30/2017 0105 hrs. local 5.2% full, 0.5 psi exhaust pt243 @ 1.98 x 10-9 torr
2 am: 5.9% and 0.5 psi exhaust pressure. PT-243 = 1.94e-9 Torr.
0300 hrs. local 5.4% full @ 0.6 psi and 1.92 x 10-9 torr
4am 5.6% full @ 0.5 psi and 1.89 x 10-9 torr
0500 hrs. local 6.1% full @ 0.8 psi exhaust. PT243 1.87 x 10-9 torr
6am 7.1% full @ 0.7 psi exhaust. PT243 1.84 x 10-9 torr
0700 hrs. local 7.7% full @ 0.9 psi exhaust. PT243 1.83 x 10-9 torr We are using a T-size GN2 cylinder which is typically filled with 300cubic*ft of GN2. So, (300cubic*ft)(28L/cubic*ft)(min/2L)(hour/60min)(day/24hour) = 2-3 days of flow
I noticed on-scale values for CP3's level for a few snapshots - could it be?
Kyle and I are headed to CP3 to see if we can blow out the newly forming passage.
Here is a trend. Looks like it started coming on scale around midnight local time. Watching it live from control room, it's fluctuating between 0.1 and 7.2 from the little bit I watched it. Heading to mid-Y now to set up N2 bottle. We'll apply 100 psi to sensing line.
The FMCS is indicating that the Y-mid is 72F at this time. Knowing that a new HVAC control system has recently come on line, as well as the knowledge that the Y-mid VEA temp had been "high" at one point last week (at the time, I had noticed increases in pump currents for all Y-mid AIPs which is typical of an increase in room temperature. I confirmed with Bubba at that time that he was aware of the increase room temp.) I decided to independently measure CP3. See attached image. This is only "interesting" and no conclusion should be made as nothing is calibrated yada yada yada.........
I wonder what's cheaper: LN2 at 35 cents a gallon or electricity at 6 cents a kWh.
According to PSI V049-1-033 (aka LIGO-C960964, attachment 4, p. 98 ff) the short CP in “iced” surface condition (thermal emissivity ~ 0.9) absorbs 250W thermal blackbody radiation from each end aperture, assuming 80K reservoir and 300K beamtube or chamber blackbody temperature. Another 102W are expected to enter through conduction, supply line loss, and local shell radiation [1].
With other factors fixed (i.e., midstation climate-controlled, pipe insulation good, etc.) the boiloff rate vs. variable beamtube temperature can be approximated as
mdot ~ P/C
~ [352W + 250W*(T_bt/300K)^4 ] / C
~ (602W + 1000W*[(T_bt-300K)/300K])/C
= 2.6 g/s @ T_bt = 0C, 3.0 g/s @ 27C, and 3.2 g/s @ 40C (32, 81 and 104 F respectively)
where mdot is mass evaporation rate (g/s), T_bt is the beamtube temperature, P is absorbed thermal power (W), and C is LN2 heat of vaporization (~ 200 J/g).
In other words, short CP LN2 consumption should increase by about 0.6 % per degree C of beamtube temperature [2].
We should install BTE thermometers as done at LLO. Just one or two per module would probably be adequate.
[1] The PSI calc assumes heat shielding and reflective liners built into the CP have very low emissivity (< .05), which may not be true any more; if the tube liner finish or cleanliness is compromised due to age, radiant heat load could be 1.5x greater.
[2] At LLO naked tube steel now tracks BTE enclosure air temperature closely [G1400433]. BTE air temp is a function of outside ambient, sun exposure and the insulating and heat storage properties of the BTE concrete, but we think both sites hit at least 49C peak in summer. Diurnal variation is about 20C p-p.
Assuming that I can find the flanges that I need, I will be reworking VBOC in the VPW and expect to have a few hours worth of work. Sheila D. is currently working at X-end. I will contact her and make a comment to this entry when I am leaving.
Made good progress towards re-configuring VBOC but there is still much more to do. 1540 hrs. local -> Kyle leaving site now. Leaving Kiwamu and Sheila to their thoughts.
It seems that the ALS laser at end X is occasionally having a sudden drop in green power, starting about 6 hours ago and getting worse. I looked through 200 days of trends of the green power and don't see anything like this. This is the same laser which has its diode current increasing, see alog 36417
I have not been able to lock ALS for a while, most of the locklosses aren't at times when this power drops, but I am wondering if the laser or SHG is the cause of the problems.
The second screenshot shows some end laser powers over an hour, you can see that at some times the IR power drops at the same time as the green power, and that the TMS QPDs also see the power drops, so I think this is a real problem not just a problem with the monitor diode.
It could be that the laser is mode hopping and we just need to change the temperature to find a mode hop free region, or maybe we need to get out a spare laser.
before I left last night, I drove to End X and power cycled the laser, at 4:59 UTC on May 29th. In the attached screenshot you can identify this by the time that the LASERDIODEPOWERMONITOR goes to 0. ALthought the power out of the laser decreased by about 3% for IR ( but only 1% for green, which should depend on the IR power^2) the sudden power drops didn't happen for 10 hours after I power cycled it.
I was able to lock ALS for about 6 minutes, but am going to head to the end station now to check for mode hopping in the ALS laser.
I was able to move the laser temperature so that the beat note moved by +-500MHz, without seeing any mode hopping.
Attached are two screenshots of two different kinds of power drops, one that only appears in green channels and one that only appears in IR channels.
I also adjusted the temperature of the doubling crystal, which was at 33.51 when I started, to get about 3% more green power out. I don't think this will fix the problem.
To prepare for swapping the laser, I checked the green irises on the input beam right before the periscope, the beam is well centered on them. I also added two irises in the IR path right after the laser before the steering mirror, using a card not a viewer. I attempted to add an irises to the green path but did not have the right base plates.
I also added and aligned 4 irises right after the laser, in the IR path the two irises are in the longish beam line before any steering mirrors, for green they are in the path after two steering mirrors before the EOM. I also put 4 dog clamps marking the location of the laser base plate. We can use the dog clamps and the irises in the IR path for placing the laser, and use the irises in the green path for the two steering mirrors right after the laser, to hopefully get us to roughly the right alignment.
