J. Kissel

Robert has left the site, and I have to head out. After the long stretch's lock loss, I needed to tweak the ETMY alignment by ~0.5 [urad] before the green WFS were able to take over. Other than that, the lock acquisition system has taken over and run with it. So far there have been two failed attempts in the same way -- it gets all the way up to the middle of the power increase (right as the CSOFT and DSOFT loops come on) and then it looses lock. Maybe the ITM OL damping loops need to be off for the lock acquisition sequence and then turned on later? Dunno.

The IFO disappears for a few days starting tomorrow. Lots of electronics need swapping. Hopefully it will be quick to return to this great Sunday's kind of locking around Thursday or Friday.

J. Kissel, R. Schofield

I'm not exactly sure when Evan left last night, but this 23 [W] lock stretch has lasted for 13+ hours thus far (guardian says the lock stretch reached ful LSC_FF at 08:50:27 UTC, which concurs with the arm cavity power -- see attached). Regrettably, Evan didn't hit the Undisturbed bit, but we should consider almost all of the hours this lock stretch entirely undisturbed until Robert goes in. At worst, the landscaping crew had been driving near the building. We've had no substantial Earthquakes, and the wind has stayed around 10 [mph].

He's delaying as long as he can, but Robert has to pack up his HAM6 shaker set up eventually, so that will likely be the cause of the lockloss if the Earth doesn't do it for us. 

C'mon galactic supernova!!

P.S. @DetChar and @CDS -- we REALLY need a way to trend the inspiral range in the control room (that doesn't require matlab or python). Sooner or later, we're going to compute the range in the front end (like LLO already does) if not!

I did some broadband intensity noise injections from about 05:04:00Z to 05:32:00Z, 2015-05-17. I also took another frequency noise coupling TF. Noise projections forthcoming.

I widened the BS violin stopband filter (FM5 in BS M2 LOCK L, 80 dB elliptic). In the course of doing MICH noise injections, I saw a wide response around the beamsplitter violin mode, as was seen at LLO (see Brett's post and the posts linked therein for a discussion). The stopband used to go from 297 Hz to 307 Hz; now it goes from 250 Hz to 350 Hz. These injections seem to indicate that the beamsplitter roll mode is also quite wide, but since this is so close to the UGF of the MICH loop I left the bandstop filter alone.

I took OLTF measurements of PRCL, MICH, and SRCL. PRCL UGF is about 45 Hz and SRCL UGF is about 25 Hz. The MICH UGF was about 12 Hz, with 12 degrees of phase (!). It seems the compensation filter from a few weeks ago is no longer necessary in the full, low-noise 23 W configuration. I'm not sure whether it's necessary at some earlier point in the locking sequence, so I've left it in the guardian for now.

We have seen for a while now that when we transition control of DARM from rf to dc readout, the fluctuations in OMC DC can be as bad as 30% pkpk. We've also seen that the interferometer sometimes loses lock either on the transition step, or on the subsequent step of switching actuators from ETMX to ETMY. For the record, one can do the final DARM stabilization steps (bringing the UGF up to 50 Hz, and engaging the boost) before transitioning to dc readout. This reduces the RIN in OMC DC to less than 10% pkpk. Then, in order to switch actuators, ramp the drive of ETMY ESD on simultaneously while ramping the drive of ETMX ESD down. I used a 30 s ramp, but I suspect we could get away with 10 s or less. I have not put this in the guardian.

I have seen the same 0.4 Hz oscillations in full lock, as Jeff reported earlier. To get around this tonight, I left the ITM oplev damping on. Removing the damping even in full lock leads to instability.

J. Kissel 

The message: I tried to improve the Bounce mode damping, I've installed some new filters; no change in the amount of damping for the bounce modes, and they haven't been cooled any more than when I got started. I've made no changes the guardian code, so in the next lock stretch my work will be erased / cleaned up (except for the new filters), and what used to be turned on turned on before today will be turned on again.

After spending an hour or so trying to get better results out of the installed bounce mode damping filters to no avail, I tried installing much more narrow band-pass filters. I did this because of Shiela's entry (LHO aLOG 18440) seemed to indicate we know the frequencies really well, and I could see what I thought was ETMX at 9.77 [Hz] interacting with what I thought was ITMX at 9.83 [Hz]. However, I found after exploring the parameter space (with gains and +/- 30 or 60 [deg] filters) with the more narrow filters, the situation got more confusing because the top mass of ETMY, ETMX, and ITMX would come in and out of showing 9.83 [Hz]. See for example the attached .pdf of the top mass vertical OSEMs for each. Yuck! After a while I convinced myself that having both IX and EX using the narrow band bp9.7-9.8 (in FM10 of IX and FM5 in EX) with a small gain of +0.1 and +0.05 and their normal -60 [deg] and +60 [deg] filters wasn't making it worse, but I couldn't get any improvement.

This mode interaction are stable / non-existent at low power, but as soon as we elevate to 23 [W], the MICH oscillation creeps in mentioned earlier today (LHO aLOG 18478) creeps in and breaks the lock.

J. Kissel

I've processed the DARM OLG TFs from Thursday (see 18443), and found that the DARM Coupled Cavity Pole Frequency seems relatively stable at 350 +/- 10 [Hz] with respect to IFO power. Admittedly, the data at 23 [W] is only over a small frequency range and is varying by ~5% (even though, as usual, I applied at 0.99% coherence threshold cut on the data points), so that's why I put a +/- 10 [Hz] on the number I quote. I think we'll either have to (a) get more DARM sweeps, (b) use Kiwamu's LSC DCCP tracker, and/or (b) get the PCAL demodulation up and running to check it this is consistent from lock-stretch to lock stretch.

The ITMs are under control for all of these data sets, and the SRC1 yaw loop is tuned as described in LHO aLOG 18442 for the latter three. For all four measurements SRCL to DARM subtraction, ("FF") was off.

IFO back up; bounce mode looks better (maybe a slightly different alignment?); 2.5 [kHz] violin harmonic is still quite rung up. 

Turns out our DMT session had just frozen on the wall -- the IFO has been locking itself for the past 8 hours!! The lock stretches have only been ~1/2 hours long though, I'll see if its these poorly damped modes.

Robert has joined the party, and he plans to do injections later in the morning, but for now I've checked the undisturbed bit at 2015 May 16 16:10:00 UTC.

J. Kissel

I came in ~10 minutes ago, and found that the IFO had just finished getting to low noise, in the LSC_FF state. ITMX's highest vertical (bounce) mode and the 4th harmonic (2.5 [kHz]) of somebody's violin mode are very rung up, but the IFO appears to be stable(-ish). I briefly turned on the undisturbed bit, but when I noticed that DMT wasn't computing the inspiral range, I found there was a state above LSC_FF, "NOISE_TUNINGS" so I've turned off the undisturbed bit, changed the nominal state to NOISE_TUNINGS, and went for it (since all it appeared to do was turn on a SRCL offset). This broke the lock after a minute or so. Hurumph.

I've now changed the nominal state back to LSC_FF, and I'm gunna leave it there is time. Once it gets back up (hopefully), I'll reengage the undisturbed bit.

Brute Force coherence report is available here for last night lock:

https://ldas-jobs.ligo.caltech.edu/~gabriele.vajente/bruco_1115716006/

Some highlights, which I find very interesting:

• What's going on with CARM, why is there coherence of basically 1 below 20 Hz? (fig. 1)
• There is coherence at the same peaks, and a bit at all frequencies with AS_A_DC_SUM: is it intensity noise? (fig. 2) This seems to be confirmed by the coherence with IM4_TRANS and ISS signals (fig. 6)
• SRCL is still a good channel to have coherence below 80 Hz. (fig. 3)
• AS_A_RF_45_Q_PIT has large coherence below 20-30 Hz, maybe we are limited by angular noise? Do we need a better balacing or centering of the ETMs? (fig. 4)
• At 17 Hz there is a small bump/line that is coherent with HPI-HAM1_BLND_L4C_HP_IN1 (fig. 5)
• Coherence with the PSL periscope is much smaller than in the past (fig. 7). However, there is still some significant coherence with IMC WFS at 200-400 Hz (fig. 8), so we're probably still suffering of some contamination due to beam jitter

Not much is happening at higher frequency, at least for broad band coherence. However there are a lot of single frequency bins with significant coherence, and I can't list all of them here (I'm too lazy). So look into the table at your favorite line frequency, you might be rewarded with some coherence.

I spent a bit of time this morning clearing out remaining SDF diffs in SUS, LSC, and ASC.  This involved the usual alog hunt followed by some clarifications from commissioners.  Things still showing diffs at the moment and why:

- We need to capture a new snap file for the OMC due to yesterday's model change which will clear out it's large SDF diff list. 

- The ASCIMC still has one diff due to a 10e-17 size change in one value.  I think this feature is due for an upgrade at some point, so until then, this 1 diff will be lingering.

- Same with ISCEX - there are 2 diffs with very small diffs which won't go away with the accept button.  Will linger until update.

- We need some assistance with clearing the PSL diffs from team PSL.

- I've not paid much attention to the bottom monitors, namely ODC and CAL.

Sheila, Evan, Stefan


- Reduced the whitening gain on both AS 36 ASC diodes. The new value is 24dB. We increase the input matrix values from 1 to 2 to compensate (for MICH and SRC1, i.e. the SRM loop).
- We tried to phase the AS_A_36 quadrants by wiggling SRM in yaw and pitch, maximizing the I signal. Not sure how meaningful that was though, because the actual signal we ended up using was Q.
- We switched the the sensor for SRC1_YAW (SRM) to 0.3*AS_A_36_Q + 0.3*AS_B_36I. This signal clearly reacted to our mysterious SRC cavity drifts, and had zero offset at the good locking point.
- We increased the power to 23W. This is the maximum power currently available. With this new SRM yaw sensor the 23W was very stable.
- The DARM offset was 14pm (or 1e-5 cts).
- This gave us a recycling gain of 38.
- Measured the open loop gain and verified the calibration.
- This configuration was stable for 1h.
- We took SRCL noise injections at: 8:46:15UTC and 8:59:40UTC (SRCL cut-off filter off and on)
- We took MICH noise injections at: 8:52:15UTC and 8:55:40UTC (SRCL cut-off filter on and off)
- Pushed the intent bit at 9:06:30 UTC.

- The inspiral range seems quite remarkably flat at 57 Mpc. A stron indication that we are now purley electronics noise limited at low frequencies.

To do tomorrow:
- Explore DARM offset: the new OMC code is ready to install, and will allow on-the-fly OMC offset tuning.
- Add the new ASC INMATRIX to Guardian.
- Add the new DARM offset to Guardian.
- Generate noise budget.

the remaining roll modes are at 13.89 and 13.98 Hz, but we don't know which is from ETMX and which is from ITMX

This is a follow up entry of LHO ALOG 17601.

A couple of days ago, the discrepancy of the response for DCPDA and DCPDB were found. This was basically caused by misadjusted filter modules for the anti-whitening filters. Some of them were using design values (like Z10:P1) and some others were just left as they had been imported from the LLO setup.

In order to correctly take the whitening transfer functions into account, the wiring of the in-vacuum and in-air connections were necessary to be tracked down. The 1st attachment shows the sufficiently detailed wiring chain for this task. Using the test data (links indicated in the diagram), we can reconstruct what the correct anti-whitening filters should be. The summary can be found below.

[Trivia for Rich: DCPD1 (Transmission side of the OMC BS) is connected to HEAD2, and DCPD2 (Relfection side of the OMC BS) is connected to HEAD1. This is because of twisted D1300369. This cable has J2 for HEAD2 and J3 for HEAD1. This twist exists in LLO and LHO consistently, as far as I know]

=======
Characteristics of the DCPD electronics chain
Complex poles/zeros are expressed by f0 and Q

DCPD A
(DCPD at the transmission side of the OMC DCPD BS)
- Preamp D060572 SN005
Transimpedance: Z_LO = 100.2, Z_HI = 400.0
Voltage amplification ZPK: zeros: 7.094, 7.094, (204.44 k, 0.426), poles: 73.131, 83.167, 13.71k, 17.80k, gain: 1.984

- Whitening filter D1002559 S1101603
(This document defines the gain not at the DC but at a high frequency. The gain below is defined as a DC gain.)
CH5 Whitening
Filter 1: zero 0.87, pole 10.07, DC gain 10.36/(10.07/0.87)
Filter 2: zero 0.88, pole 10.15, DC gain 10.36/(10.15/0.88)
Filter 3: zero 0.88, pole 10.20, DC gain 10.36/(10.20/0.88)
Gain: “0dB”: -0.051dB (nominal), “3dB”: 2.944dB, “6dB”: 5.963dB, “12dB”: 11.84dB, “24dB”: 24.04dB

DCPD B
(DCPD at the reflection side of the OMC DCPD BS)
- Preamp D060572 SN004
Transimpedance: Z_LO = 100.8, Z_HI = 400.9
Voltage amplification ZPK: zeros: 7.689, 7.689, (203.90 k, 0.429), poles: 78.912, 90.642, 13.69k, 17.80k, gain: 1.983

- Whitening filter D1002559 S1101603
CH6 Whitening
Filter 1: zero 0.88, pole 10.13, DC gain 10.41/(10.13/0.88)
Filter 2: zero 0.87, pole   9.96, DC gain 10.40/(  9.96/0.87)
Filter 3: zero 0.88, pole 10.15, DC gain 10.41/(10.15/0.88)
Gain: “0dB”: -0.012dB (nominal), “3dB”: 2.982dB, “6dB”: 6.007dB, “12dB”: 11.87dB, “24dB”: 24.04dB

=======

Now we put these transfer functions into the model and check if we can reproduce the observed relative difference (Attachment 2). In deed, the measurement is well explained by the model below 30Hz where the measurement S/N was good. As we saw in the previous entry, the difference of the DCPDA and DCPDB after the whtening compensation is 20% max. Note that further inspection revealed that this 20% difference is, in fact, mostly coming from the difference of the preamp transfer functions rather than the miscompensation.

So this was the relative calibration between DCPDA and DCPDB. How is the compensation performance of each one? The 3rd attachment shows how much of current we get at the output as H1:OMC-DCPD_A_OUT, H1:OMC-DCPD_B_OUT, and H1:OMC-DCPD_SUM_OUT, if we give 1mA of photocurrent to DCPD_A, DCPD_B, or both (half and half). Ideally, this should be the unity. The plot shows how they have not been adjusted. For the our main GW channel we take sum of two DCPDs. The individual deviations were averaged and thus the sum channel has max 10% deviation from the ideal compensation. This shows up in the GW channel.

=======

So let’s implement correct compensation. Basically we can place the inverse filter of the each filters. The preamplifier, however, includes some poles and zeros whose frequency are higher than the nyquist frequency. Here we just ignore them and assess how the impact is.

The result is shown as the 4th attachment. Upto 1kHz, the gain error is less than 1%. This increases to 5% above 3kHz.  The phase error is 7deg at 1kHz. This increases to 20deg above 3kHz. These are the effect of the ignored pole/zeros. Note that these are static error. In fact, the phase error is quite linear to the frequency. Thus this behaves as a time delay of ~18.5us. Since the phase delay at 100Hz is small, the impact to the DARM feedback servo is minimal. For the feedforward subtraction, however, this might cause some limitation of the subtraction performance. In practice, we measure the coupling transfer function in order to adjust the subtraction, in any case. Therefore this delay would not be a serious problem.

The filter bank to implement the new compensation was already configured. The filter file is attached as foton_DCPDfilters.txt.