Team TCS

After having the ring heaters on last night we put in the CO2 pre loading estimates suggested in alog 43979.

After testing out this initial setting and various other differential lens settings  (Please see attached time series (ignoring the high frequency noise of the Michelson losing lock)) we observed the following: 

1. The contrast defect estimate is about .4% with mich bright around 7200 counts and mich dark around 32 counts.
2. The most optimal contrast defect setting is at about 10 microdiopters positive lensing from the cold state for ITMX and 3 microdiopters of positive lensing from the cold state for ITMY. 
3. When reaching reaching the dark michelson we sill observed some higher order mode content on the ISCT6_AS_AIR camera (please see attached image). 

TVo, Danny, Hang

We did a quick Finesse simulation to see how much contrast defect we should expect at the simple mich configuration, as a function of differential ITM thermal lens. Here the contrast defect was defined as the AS power at MICH dark/ power at MICH bright. 

Please see the first attached figure. We expect around 0.2% contrast defect at 10 microdiopter (100 km differential ITM lens). This should roughly serve as a requirement on TCS. 

On the other hand, to get good SRM alignment signal & small SRM detuning (which affects the low freq DARM response), we need to control the differential lens to <~ 10 microdiopter. 

The second (no diff lens) and third figures (100 km diff lens) show how the AS72 signal changes. 

The last figure show the simulated DARM TF at different levels of thermal lens. 

To make space for the new h1cdsrfm machine (long-range dolphin), I moved h1build up by 2U in the MSR rack. After h1build was restarted, I restarted all the Slow Controls SDF targets.

TITLE: 09/13 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Commissioning
INCOMING OPERATOR: None
SHIFT SUMMARY:

Day mostly dominated by TCS work
LOG:

15:30 Vanessa out to LVEA - cleaning

16:30 Nutsinee out to HAM6 area - ISCT6

16:59 Vanessa out to EX

17:15 Haocun out to LVEA - w/Nutsinee

17:33 Nutsinee out

17:34 Vanessa out to EY

17:41 Mark, Tyler, and Chris out to work on the outside of EX sealing bat-holes.

18:37 Marc and Dave ou to LVEA - cabling from H2 building into LVEA

20:00 Commissioning meeting

21:13 Mark, Tyler, and Chris back from EX

21:18 Marc and Gerardo pulling cables back into H2 building from LVEA - Gerardo into LVEA; Marc into H2 building

21:25 Gerardo back

21:38 Gerardo out to LVE to valve in neg pumps - aLog

22:11 Nutsinee, Haocun, and Terry out to TCSY area

22:37 Terry back - Nutsinee and Haocun still out

{Gerardo, Chandra}

After regenerating the three corner station NEG pumps on Tuesday, we observed the local housing pressures at NEG 1,2,3 for a couple of days before valving into main volume. While valving in (waiting 10 minutes between each NEG), we scanned the RGA and observed no noticeable change in AMU peaks. The total pressure decreased slightly.

Time stamps of each for RGA correlation (all local times): 

1. 2:45pm valved in NEG1
2. 2:56pm valved in NEG2
3. 3:06pm valved in NEG3

Gerardo noticed gappy flanges on the BT side of the NEG isolation gate valves and Kyle does not recall spraying He directly at these on Tuesday when we "fire hose" leak checked the corner. These three units are near the RGA, where we observed the most notable increase in He signal while leak checking on Tuesday, so we would like to leak check these again.

The OMC whitening chassis has been modified (42361) so that the second whitening stage is now a low pass filter.  The idea is that we can engage this if the violin modes get rung up so badly that we need to reduce the analog gain to avoid commissioning. Jeff K installed the compensation filters yesterday, and now I have modified the OMC_LOCK guardian so that it can be used to engage this low pass or remove it, and so that the add and remove whitening states will only change the stages that are still whitening stages. 

I've tested this without the OMC locked and it works, but it would be a good idea to test it with the OMC locked before trying it at DC readout to make sure everything is OK with the filter and it's compensation. 

Chandra asked about pod pressures, so here they are. I thought this was a FAMIS task, but I don't remember seeing it in a while.  Maybe we put a hold on it during the vent? Searching for "pod pressures" gave me nothing about ISI pods at least as far back as the end of May, too lazy to look harder.

The corner BSCs and ETMY show some trend down at the end of July of about .2 kpa that doesn't seem to show up on the HAMs or ETMX. But otherwise I don't think any of the pods are leaking. 

Daniel, Nutsinee

TTFSS

Instead of a phase-frequency discriminator we now have a IQ demod inside the new TTFSS. It's working, but not quite as it should. The input plus/minus switch needs fixing, UFG looks miserable (33kHz zero crossing, used to be 300kHz and can be pushed to 500kHz), need more gain. We are now lock our pump laser to the OPO error signal with this TTFSS (S1812013). 

Here's a transfer function taken at the common path.

Pump refl PD saturation

Looking at the RF mon signal (signal before going to demod board), not only we saw 80MHz and 160MHz (2 omega) but also 

35.3MHz, 44.4MHz, 70.6MHz, 115.4MHz, and 199.9MHz

35.3 MHz is likely our 35.5MHz modulation frequency for the SHG (I used peak search button on the Agilent not sure how accurate the frequency is). Though we didn't expect it to be as high as the 80MHz. From my note we were driving the EOM with 1.5Vpp, the EOM modulation depth spec is 15mrad/V. So total is 22.5 mrad/2 ~ 0.01 modulation depth. This is Newport 4004. Could it be that our carrier is so large that .01 of that is equal to 80MHz sb magnitude? SHG linewidth for green is 30MHz. 35.5MHz will mostly reflects back and goes to OPO. SHG is locked on transmitted red.

Here attached a photo from Agilent screen, left is when the cavity is unlocked (but hanging on resonance), and right is when it's locked. This was taken with ND1 filter in front of the PD. RF mon has 23 dB attenuation, ND1 counts for 20 dB. So before when we weren't locked at the peak the PD used to have -30+23+20 = 13dBm of 80MHz. 

Don't know where the rest of the lines came from. They're not that small either. All of these probably contributed to our refl PD saturation problem.

As a by-product of out recent investigations to understand the CHARD angular response, we measured again the angular response of all four test masses, by driving the L2 stage from the ISCINF_P and ISCINF_Y channels and reading out the motion using the optical levers. We also measured the cross coupling, i.e. yaw to pitch (Y2P) and pitch to yaw (P2Y).

Measurements

In summary: the ETMs have reasonably low P2Y and Y2P cross-coupling, but the two ITMs show quite high Y2P cross coupling. In particular, ITMX shows a 1:1 coupling of yaw to pitch around 1.5-2.0 Hz.

Cross-coupling fits

If we want to compensate for those cross-couplings in the suspension drive, we need to compute ratios like Y2P/P2P and P2Y/Y2Y for each suspension and fit them. The two plots below show all such terms for ITMs and ETMs

As already commented, the ITM yaw to pitch coupling is larger than any other cross coupling.

I fitted all the ratios above with MATLAB and vectfit, to obtain the filters listed below, in foton format. The fit plots are also attached to this entry.

Filters have been uploaded to the suspension L2 DRIVEALIGN, but not engaged. The gain in the P2Y and Y2P filter banks should be -1 for all of them.

TITLE: 09/13 Day Shift: 15:00-23:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Commissioning
OUTGOING OPERATOR: None
CURRENT ENVIRONMENT:
    Wind: 6mph Gusts, 4mph 5min avg
    Primary useism: 0.02 μm/s
    Secondary useism: 0.14 μm/s
QUICK SUMMARY:

For the TCS+ISC team

With the success of increasing the power last night to about 17 Watts input, the chance to make an absorption measurement for the ITMs became a reality and we were able to use both Hartmann sensors to extract a spherical power which we fit to a Comsol model, this technique is the same as Aidan's and Georgia's method. 

ITMX absoprtion: 206 +/- 1.8 ppb

ITMX_RH tuning for 50 Watts:  0.44 Watts Total

ITMX_CO2 pre-loading: 1.03 Watts Central

ITMY absoption: 454 +/- 1.5 ppb

ITMY_RH tuning for 50 Watts: 4.19 Watts Total

ITMY_CO2 pre-loading: 2.27 Watts Central

Plan for tomorrow:

- We'll plug in these values for the ring heater tonight, by the morning they should be relatively close to thermal equillibrium.

- Turn up the power of the CO2 lasers, and measure the simple michelson contrast defect.  If it works, we should be able to lock at 2 Watts.

Details

I was hoping to figure out a way to estimate the uncertainty in this measurement but it's a little bit more involved than I thought,  the method is essentially using a least squares fit to get the coefficients for the exponential decay of the thermal lens after losing lock, but the residuals seem to be pretty good which leads to about a 1% accuracy which should be taken with a large grain of salt because the absorption estimate is made by interpolating the scale of the thermal lens' exponential decay and then dividing the coefficient by the arm power in Watts.  However, the arm power is an estimate as well with its own error bars which I'm still trying to figure out.

P_arm = 17 Watts * 45(PRC_Gain) * 0.5(BS) * 282(Arm_gain)

Once we got estimates for the absorption (1st and 2nd attachment), we re-ran the calculations which gave ring heater settings to compensate for the 50 Watt nominal power based off the absorption.  ITMY ring heater requirement is particularly high because of excess absorption.  Also to pre-load the system, the CO2 laser settings are also calculated (3rd and 4th attachment).

We were able to modify the ARM ASC plant by digitally adding the Sidles-Sigg torque.

Specifically, at 2 W input power, we are able to digitally modify the plant to look like a 10 W one, and at 10 W, we can reduce the plant back to the 2 W one. This further means that we should be able to remove the power-dependent part of the ctrl filter, and designing a single ctrl filter should be sufficient.

Please see the first attached plot.

Here the red trace was the DHARD YAW OLTF measured at 2 W input power without digital compensation.

The blue trace was the DHARD YAW OLTF at 2 W input power but with a digital Sidles-Sigg torque added so that it looked like a 10 W plant. Note that the sus resonance was shifted digitally from 1.4 Hz to 1.5 Hz. (We increased the overall loop gain to make the UGF 6 Hz for this measurement.)

The green trace was the OLTF measured at 10 W input power with the digital compensation to reduce the plant back to the 2W one, as the resonance peak was shifted back to 1.4 Hz. (As we power up the optical response decreased a bit thus the UGF reduced to 4 Hz; yet this is only a dc factor and can be tuned easily. Also we do not need a perfect subtraction here. With 10% error we can already reduce the 50 W plant to a 5 W one which we can control without modifying filters. The DC gain also provides us a measurement of the )

As a reference, in the second plot we also show the matlab modeled DHARD YAW OLTF. My sus model was just the default one so the resonance frequencies did not match the real LHO quads exactly, but the point is that from 0 W to 10 W (50 kW per arm circulating power), we should expect the secondary suspension resonance at 1.4 Hz to be shifted to 1.5 Hz, as illustrated in the measurements.

==================================================================================================================

Details:

We currently do not have the path dedicated for this digital compensation. Thus we borrow the DC 5 filter bank for this compensation. The same error signal for DHARD Y ctrl was sent here and the output goes to the same output matrix as DHARD Y ctrl. The details of the filters was detailed in LHO:43849. We updated the optical response FM5 [rad/ct] by increasing it by a factor of 3. Changing the 2 W plant to 10 W we put a DC gain of +1. To compensate for the 10 W plant to make it back to 2 W we put a gain of -0.8 (due to lose in the optical gain as we increased power; again this is just a DC factor and tuning it to 10-20% accuracy should be much easier than designing different ctrl filters at different input powers).

Tonight we stayed locked at almost 15 W for an hour to get a nice steady state thermal lens measurement for the Hartmann (with limited success for ITMX, see log post 43981)

We lost lock in the same manner as last night, with the OMC suspension saturating. Hang and I had a quick look at the witness sensors for many optics and can see that over the duration of the lock PRM and SR1 are shifting by roughly 1 urad.

This attachment has many plots: top left is X arm transmission and pop rf18, which increase as we acquire but then decrease as we sit at 15 W and the thermal lens increases. Bottom left shows the test mass positions as measured on the op levs. Right plots shows PRM, PR2, BS, SR2, SRM witness monitors on the same timescale.

We reduced the ASC loops gain by a factor of 4 in the MICH_DARK_ALIGN state in ALIGN_IFO guardian and now it seems to be working again. We should be able to use it for again for doing simple mich contrast defect measurements.

We took a DHARD YAW measurement at 2 W at GRD state DARM_TO_DC_READOUT. See the first attached plot. The UGF was 4 Hz yet it seemed fine to increase it up to 6 Hz as we still had a phase margin of 50 deg. Thus we increased the DH_Y gain from -40 to -60. Similarly we increased DH_P gain from -30 to -50 to make the loop more robust for power up.

On the other hand, we noticed some oscillation in CH_P at 6 Hz (and also seen in many other loops as shown in the second plot; yet CH was the only loop had a UGF high enough at this point that could potentially cause gain peaking). The oscillation was present even before we increase the DH loops gain. We tried to reduce CH P gain yet the gain peaking got worse. Instead we increased CH_P gain from 0.3 to 0.4 and it slightly reduced the 6 Hz oscillation. Also turning up the DH loop gains also mildly helped. See the third plot. The green trace was before all the gain increasing and the red one was after we increased CH_P from 0.3 -> 0.4, DH_P from -30 -> -50, and DH_Y from -40 -> -60.

Will try to get a high-resolution CH OLTF once we have chance.

[Sheila, Hang, Gabriele]

In brief, the L1 LOCK gain in ITMX was 10 instead of 1 (as it used to be during O2). Therefore the reallocation of pitch and yaw to M0 was ten times stronger, and it modified the suspension angular response as seen by the ASC at the lower stages. This created a crossover at about 1 Hz that produced the right-half-plane zero we measured in CHARD and DHARD. This wrong gain was there only in ITMX. We reverted it to 1 as it was during O2. The gain was changed from 1 to 10 on Saturday June 2nd 2018 at about 21:55 UTC. Nothing relevant written anywhere in the elog.

The plot below shows the yaw to yaw response of ITMX, measured with the optical lever, in a few different configurations.

• Blue: M0 reallocation on, as it has been for the past 3 months (ten times stronger than it should be). This transfer function deviates from the expectation (and from other suspensions) below a few Hz. You can also note the large phase rotation at about 1.8 Hz, which corresponds to the non-minimum-phase zero
• Green: M0 reallocation off, this is the expected response of the suspension, and there is no weird phase rotation
• Red: M0 reallocation, turning the L1 lock gain back to 1 from 10. Note that we are not using L1, but this is a pass-through gain
• Brown: M0 reallocation plus a 0.6 Hz Chebyshev low pass filter, that Hang designed to reduce the impact of the reallocation. The design is based on the suspension model, which seems to match reasonably well the measurement (at least qualitatively)

The expectation is that this will fix the DHARD/CHARD plants (43844)

We implemented the low pass filters in all test mass M0 stages, since it gives us some more phase and gain margin at about 1.8 Hz, where there is still a zero.

The story of how we found this problem is long and convoluted. A brief summary below:

• we noticed that there was a large coupling of yaw actuation to pitch motion for the two ITMs, much larger than the ETMs
• we wanted to measure and fit it, so we repeated optical lever measurements of all angular transfer functions of all four test masses
• we noticed that the ITMX pitch to pitch was clearly different from the others, and we traced the difference w.r.t. ITMY to the fact that M0 was on for ITMX and off for ITMY
• we then took a ITMX measurement with M0 reallocation off, and found that ITMX was now behaving
• Hang added the low pass filter in the M0 path, and we found that the yaw transfer function was much better
• We finally found the wrong gain looking at the conlog

(In this post I mean "railing" as "parked at the maximum actuation range", where "oscillation" means "PZT is fighting the EOM")

The FSS Fast Path has been railing recently, preventing the FSS from locking.  (Pic 1)

It seems as though FSS_OSCILLATION does not in fact monitor the FSS RMS, but triggers on a threshold.  When the FSS voltage would pass a certain value, the PSL_FSS guardian would send it to state FSS_OSCILLATING.
This state brings the common gain down to -10 dB, then slowly back up to wherever it started.  This works great for stopping oscillations, but not for permanent rails.

I modified the PSL_FSS code to increase the oscillation threshold from 0.6 to 3 V for five seconds whenever the fast voltage is railed at greater than 10 volts with extremely low RMS.  This should allow the FSS loop to close and bring the temperature within range.  Railing hasn't happened again, so I haven't had a chance to test the state, but it does still suppress actual oscillations.