Since we're getting into the regime of higher power operations, we need to start thinking about TCS from a higher-order mode standpoint (higher than just spherical power/defocus). The current CO2 laser central heating and the ring heater provide almost 100% spherical power correction to the optic. The SELF heating, however, still produces a partly non-spherical lens from the absorption of the Gaussian beam.

These residual non-spherical lenses will increasingly cause problems as we go to higher power.

I've run a quick analysis of how much TEM00 is scattered AFTER the nominal spherical lens is corrected. This is purely for the substrate thermal lens (not for surface deformation, which is about 1/12th of the size).

For this model:

• I've assumed between 200ppb and 300ppb absorption for each optic.
• Power level in the arm:
  • O1: ~100kW
  • O2: ~170kW [check this]
• The total substrate lens (Optical Path Distortion [OPD]) was calculated in COMSOL
  • then scaled by the absorbed power and
  • scaled by a factor of 2 to account for the round-trip through the ITM.
• The OPD then has the nominal spherical power removed to yield a residual thermal lens.
• The TEM00 (beam radius = 54mm) scatter from the residual thermal lens was calculated.

The scatter is calculated in the following way:

w0 = 54E-3;

k = 2*pi/1064E-9;

E1 = exp(-(r/w0)^2)

OPD1 = uncompensated lens (dobule-passed)

S0 = best fit defocus (spherical power) to uncompensated lens

OPD2 = OPD1 - S0/2 * r^2   % the residual thermal lens after spherical power is removed

E2 = E1*exp(i*k*OPD2)

OL = sum(r*conj(E1)*E2)*sum(r*E1*conj(E2)) / [ sum(r*E2*conj(E2))*sum(r*conj(E1)*E1) ]   % overlap between pure gaussian and gaussian with residual thermal lens added as phase front error

scatter = 1- OL

The resulting plot shows the fractional scatter versus the absorbed power.

For reference, we'd really like to keep the round-trip scatter to 0.1% or less. The next step should be reviewing a detailed SIS model to try to refine these numbers. 

Conclusion: we should consider a more appropriate CO2 heating beam shape for O2B.

During the power outage recovery, Nutsinee had difficulty getting the HEPI Pump Servo to not fault out in the corner station.  We traced that to the level continuing to trip so she and Sheila lowered the level sensor on the reservoir.  The servo managed fine after that.  This morning the level is less than 1/16" lower than my latest level reading from last Tuesday.  It is cooler today and this fluctuation is within typical temperature fluctuations. So, I don't suspect a leak although I will visually check all the lines for signs of a leak tomorrow.  Neither do I suspect an accumulator leak although the abrupt pressure shift from the outage may have jarred the gas charge side schraeder valve.  I would be more inclined to suggest an air buble that was caught somewhere finally managed to burp its way out lowering the volume level.  And finally, since the level sensor setting is invasive and blind, I likely had it set very close to the trip point and Nutsinee suffered the worst of it.

Attached cooling circuit related trends since the chiller swap and power outage.

model restarts logged for Sun 11/Sep/2016
2016_09_11 00:29 h1fw2
2016_09_11 01:11 h1iopoaf0
2016_09_11 01:11 h1oaf
2016_09_11 01:11 h1pemcs
2016_09_11 01:11 h1tcscs
2016_09_11 01:12 h1calcs
2016_09_11 01:12 h1ngn
2016_09_11 01:12 h1odcmaster
2016_09_11 01:12 h1susprocpi
2016_09_11 02:19 h1fw2
2016_09_11 04:11 h1fw2
2016_09_11 06:02 h1fw2
2016_09_11 07:56 h1fw2
2016_09_11 09:52 h1fw2
2016_09_11 13:36 h1fw2
2016_09_11 15:24 h1fw2
2016_09_11 17:12 h1fw2
2016_09_11 18:59 h1fw2
2016_09_11 20:52 h1fw2

test frame writer fw2 has been unstable since power outage. Looks like h1oaf0 needed a restart to clear stuck DACs.

model restarts logged for Sun 11/Sep/2016

Power outage, all machines were restarted.

model restarts logged for Fri 09/Sep/2016
2016_09_09 10:59 h1fw1
2016_09_09 11:11 h1psliss
2016_09_09 11:12 h1broadcast0
2016_09_09 11:12 h1dc0
2016_09_09 11:12 h1fw0
2016_09_09 11:12 h1fw1
2016_09_09 11:12 h1fw2
2016_09_09 11:12 h1nds0
2016_09_09 11:12 h1nds1
2016_09_09 11:12 h1tw0
2016_09_09 11:12 h1tw1
2016_09_09 15:32 h1calcs
2016_09_09 15:32 h1iopoaf0
2016_09_09 15:32 h1ngn
2016_09_09 15:32 h1oaf
2016_09_09 15:32 h1odcmaster
2016_09_09 15:32 h1pemcs
2016_09_09 15:32 h1tcscs
2016_09_09 15:33 h1susprocpi

fw1 bios settings were recorded. New PSL ISS code with DAQ restart. h1oaf0 restarted to clear stuck DACs.

Sheila, Terra

Tonight we saw a new PI mode ring up due to the recent increase of ITMY ring heater power: MODE 9, 14982 Hz. Damping was successful. 

Increasing ring heater power shifts the optical beat note peak (between TEM_0,0 and TEM_0,3) down in frequency, increasing the frequency overlap with the 15000 Hz and 15070 Hz mechanical mode groups (while decreasing the overlap with the 15500 Hz mode group). MODE 9 rang up at 35 W about 3 hours into a 50 W then 35 W lock. It damped with no phase and -1000 damping gain. 

I've added active damping for MODE 9 to the SUS_PI guardian.

Evan, Sheila

Since the power reduction from 50W to 35W at about 2:25 UTC Sept 11, we have been sitting with the IFO mostly undistured other than changes to PI damping.  We have not been in low noise (no ISS second loop, we are still on the high noise ESD driver), but we have seen some hugh noise in DARM that looks like a fringe wrapping shelf.  We reduced the power again at 3:46 UTC, and the noise got better and went away after 10s of mintes.  

We don't see anything in our LSC loops or main ASC loops that had a similar shape, do the detchar fringe wrapping/scattering tools offer any clues? Especially at the beinging of this time window, there is a lot of noise in DARM (a high frequency comb) which is due to a rung up PI mode. 

Evan H., Jenne, Matt, Kiwamu,

We locked the interferometer at 50 W with the latest ring heater setting (RHX = 0.5 W, RHY = 2.5 W, 29588). So far the interferometer has been locked at 50 W for roughly 1 hour.

P.S. we have now having difficulty damping a PI mode (mode2, 15520 ITMX) and decreased the PSL power back to 30 W.

[DRMI lock on POP sensors]

As reported yesterday (29601), we had a difficulty in switcing the sensors from the 3fs to 1fs earlier today (29603). In the end, I manually executed the sensor switch process one by one and for some reasons this was successful. I then measured the open loop transfer functions of the DRMI LSC degrees of freedom, but they looked OK. See the first attachjment. This may be due to that we did not wait for long enough time to let the new CO2 setting settle (29603) for lock acquisition ([CO2X, CO2Y] = [500 mW, 1000 mW]). Not sure.

Also, in lock acquisition, I manually kept aligning PR3 when the interferometer was at ANALOG_CARM in order to maintain the lock.

[CO2 tuning at 50 W]

This is not well tuned, but the below is an OK CO2 tuning for 50 W which gave us a 30% imbalance at the AS port OSA.

[CO2X, CO2Y ] = [300 -400 mW, 0W]

This time I did not spend time for tuning the dCO2 at 20 W or 40 W. Maybe I should have done that to collect more data points.

[PIs]

MODEs 17 and 27 needed a sign flip. Mode 27 seems tricky -- every time when it rang up we needed to flip the control sign.

Matt, Terra (in spirit)

I have updated the h1susprocpi model to fix a few minor problems.

• the RMSMON was at the IWAVE output, which can be misleading, so I moved it to the BP output (see first image)
• the PLL phase was in the PLL loop, but the only real impact of changing the phase was to rotate the output phase (SIG_OUT) relative to the input signal.  In-loop phase rotation can disturb the loop, so I moved the phase to the output.  This will allow for quick phase adjustment even with a low UGF PLL.  (It also flips the sign of the phase change relative to before, but the phase values are not set yet so this shouldn't matter.)  The PLL screen should be updated to reflect this!
• OSC_OUT should now have an amplitude of 1, as should SIG_OUT when the amplitude scaling is unity.  The amplitude was 2 before, so this simply divides both outputs by 2.
• changed amplitude scaling maximum from 1 to 10 ("Saturation3" in the second image).  This will allow for lower gain settings in the PI DAMP filters, which gives smaller signals when the PLL is not locked and makes the gains more similar to the ones used for direct BP damping (usually 1k to 10k).  Damping gains should not be greater than 10k, since an input of magnitude 10 will give 100k, which is close to DAC saturation.
• added amplitude scaling offsets before and after amplitude scaling saturation (AMP_PRE_OFFSET and AMP_POST_OFFSET around "Saturation3", see second image).
  • The reason for the PRE offset to is allow the output gain to go to zero (to inject little noise), even though the output of AMP_FILT is positive definite.  (Generally, the PLL_SIGNAL_GAIN should be set so that the noise driven AMP_FILT_OUT is about 0.1.)  Since this offset should always reduce AMP, it is subtracted from AMP (i.e., the value should be about +0.1).
  • The POST offset is there in case we want to make an Amplitude Lock Loop (ALL) which keeps the input signal just about the noise floor so that the PLL can stay locked and track the mode.  A small POST offset (also subtracted from the gain) will allow the gain to go negative for small values of the amplitude, thereby driving up the mode.  This has not been carefully thought out, but it seems like it should give a stable ALL which just keeps the modes above the noise.  The POST offset should be in the rage 0.0 to 0.1.

In terms of channel names, the impact of this model change should only be the addition of AMP_PRE_OFFSET and AMP_POST_OFFSET.  I have only updated the models (i.e., I have not rebuilt the FE code, etc.).  The corresponding work permit is 6149, and this change should go in with the next DAQ restart, or on Tuesday (which ever comes first).

Matt, Sheila, Stefan, Kiwamu,

We tried locking the interferometer with the latest ring heater setting (29558) tonight. The furthest point we got tonight was ANALOG_CARM.

Locking up to that point was OK with [CO2X, CO2Y] = [500 mW, 1100 mW]. However, we think CO2Y needs fine tuning to minimize the contrast defect. Once CO2Y is fine-adjusted, we wil study what is going on in the DRMI_ON_POP state.

A few minutes after 8 UTC, CO2Y tripped. 

We restarted all the models on OAF.  When I went to the chillers, only CO2Y had a low temperature warning, but there was no low temperature fault after clearing this and the chiller stayed powered on (normally after the warning is cleared the fault appears, when the fault is aknowledged the chiler shuts off.)  The laser also did not have a temperature fault when I went to turn it on, although it was off.  

Nutstinee, Stefan, Matt, Sheila, (Hugh, Dave, Jim, others on the phone)

We have basically recovered from this morning's outage about an hour ago.  There were a few things worth noting in additin to what people wrote earlier:

• I don't think that the analog camera switching is working, the images are displayed fine but we can't change which camera is on which monitor. 
• Nutsinee with instructions from Hugh tried to bring the corner HEPI pump station back but it tripped several times.  Hugh explained that this was probably because the fluid level in the reservoir was a little low for some reason, so Nutsinee and I losened the collet on top of the resevoir and lowered the sensor about a half a centimeter further into the resevoir.  After this the servo came on fine. 
• We called Jim W because the BRSs were off causing the seismic configuration manager to get stuck; it has been a windy afternoon and the optics were moving more than usual.  Jim pointed us to the good instructions in T1600103, (including a pointer to this in the power outage recovery document would be helpful).  This worked fine for Y, but we were stumped by not knowing the password for X.  Matt and Stefan were able to reset it by driving to the end station.  
• Before the BRSs were back on, we tried to reduce the motion of the optics by choosing useism from SEI_CONFIG.  This caused all the corner station ST1_CONF guardians to go into error.  To get out of this situation we requested down and reloaded.  This cleared the errors, but apparently down is not conected to the rest of the graph.  At the end of this somehow the corner station blends ended up half switched between 90mHz and 250 mHz (90mHz blend names outlined in blue for some DOFs, but staying blue and not changing.)  After calling Jim again I set the blends to 250, checked that all corner station sensor correction was off, and then manually took the guardian to the state BLEND_Quite_250_SC_None. I set each of these guardians to manual, and turned on sensor correction by changing to BLEND_Quite_250_SC_Windy.
• The highlight of the day was the new relay reset for in vacuum high voltages.  

I set up a magnetometer by the high voltage input of the OMC piezo controller as another “shot in the dark” investigation of potential sources of blip glitches. I previously set up a magnetometer by the HV inputs to the ETMY ESD. I have been focusing on HV, because of the correlation between a class of blip glitches and low-humidity that Paul, Jordan and I found (https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=28534). 

In related news, John W. and Bubba have started commissioning humidifiers. I suggest that we humidify VEAs/LVEA, electronics bays and MSR when the relative humidity drops below 7% inside of the buildings. Based on the high correlation between low inside humidity and low outside temperature last year (heating dries), an alternative that would give nearly the same threshold would be to start humidifying when the maximum outside temperature is going to be below freezing for more than one day in a row.

Channels

ETMY ESD HV: H1:PEM-EY_ADC_0_08,12,13_OUT_DQ

OMC piezo HV:  H1:PEM-CS_ADC_4_28_2K_OUT_DQ

Robert

I have all the front end computers (except mid-pem) running. IRIG-B excursions were minimal.

We ended up power cycling all dolphin'ed frontend computers and IOC chassis.

DAQ rode through the outage.

Stefan, Matt, Kiwamu,

This is a quick note. We started another TCS test in this afternoon with the new ring heater setting (RHX = 0.5 W and RHY = 1.5 W, 29557).

With a 20 W interferometer, the CO2 settings that balances the 45 MHz upper and lower sidebands were found to be at almost where we thought it should be. Then we increased the PSL power to 40 W to study the interferometer absorption as a function of the PSL power. With the sideband imbalance adjusted, the lock seems more stable at 40 W.

[good values that balances the 45 MHz sidebands]

• At 20 W
  • [CO2X, CO2Y] = [650 mW, 400 mW]
• At 40 W
  • [CO2X, CO2Y] = [650 mW, 400 mW]
  • [CO2X, CO2Y] = [500-600 mW, 200 mW]
  • [CO2X, CO2Y] = [400 mW, 0 mW]

These settings seem to give us a small imbalance within 10%. Also, lock acquisition was done with [CO2X, CO2Y] = [500 mW, 700 mW] in which I did not adjust the differential CO2 settings.

By the way, the attached trends show how the aberration changed when we went from 20 to 40 W. Notice that ITMY show relatively high aberration values (e.g. cylindrical, coma, spherical aberration) compared to the ones for ITMX. The last attachment is a StripTool showing the HWS outputs. A jump in the middle is due to us increasing the PSL from 20 to 40 W. After we sat on 40 W for approximately 1 hour, we started decreasing the common TCS and kept adjusting the differential CO2.

After finding that the ISS 2nd loop pitch and yaw signals are swapped, and inverted wrt im4 trans (my first look at qpds downstream of the IMC, alog 29542), I've looked at four qpds downstream of the IMC for signal consistancy - IM4 Trans, ISS 2nd Loop, POPA, and POPB.

What I found:

• plot 1: im4 trans yaw, iss2 pitch (inverted), popa yaw, and popb yaw - consistant
• plot 2: im4 trans pitch (doubled), iss2 yaw (inverted), popa pitch (inverted), and popb pitch - first 3 consistant, popb pitch is not a good match
• plot3:  im4 trans pitch (doubled), iss2 yaw (inverted), popa pitch (inverted), and popb pitch (inverted) - first 3 consistant, popb pitch (inverted) is closer, but still not a good match
• plot4: im4 trans yaw, iss2 pitch (inverted), popa yaw, popb yaw, and popb pitch - popb pitch is a pretty good fit for yaw

I wrote a simple interface to download data from the Tektronix network/web-enabled oscilloscopes (TDS3034B and (presumably) the like).  You can use it to download a single data file, or download data periodically:

jameson.rollins@opsws8:~ 0$ readlink -f $(which tektronix)
/opt/rtcds/userapps/trunk/sys/common/src/python/tektronix.py
jameson.rollins@opsws8:~ 0$ tektronix -h
usage: tektronix [-h] <command> ...

Interact with Tektronix oscilloscope over the network.

positional arguments:
  <command>
    download  Download data from scope periodicially

optional arguments:
  -h, --help  show this help message and exit

Use environment variable HOST to specify network address.
(default: 10.22.10.51)
jameson.rollins@opsws8:~ 0$ tektronix download -h
usage: tektronix download [-h] [-c CHANNEL] [outdir] [count] [period]

Download data from scope periodicially

Data will be saved in a two-column (time,voltage), comma-separated CSV
file named "<time>.dat", where <time> is an ISO-formatted timestamp of
the local time of the data snapshot.

NOTE: data takes about 30 seconds to download, so period should not be
less than 30 seconds.

positional arguments:
  outdir                directory to store timestamped data ['.']
  count                 number of downloads (0=Inf) [1]
  period                time between downloads in seconds [60]

optional arguments:
  -h, --help            show this help message and exit
  -c CHANNEL, --channel CHANNEL
                        channel to download ['ch1']
jameson.rollins@opsws8:~ 0$ tektronix download
output directory: /ligo/home/jameson.rollins
fetching 2016-09-09T11:16:31.747247.dat ...done (in 29.959158s).
jameson.rollins@opsws8:~ 0$