TJ, Camilla WP12277.

Started the work done in 81734 on the EX HWS at EY. Swapped the fiber collimator to a CFCS11-A adjustable SMA fiber collimator. Still need to swap to a 50um fiber, remove HWS-L3 and realign. SLED left off. 

Repeated 82151, with H1:IOP-OAF_L0_MADC{2,3}_TP_CH{10-13} 65kHz channels on CO2Y. WP# 12261.

Plots attached of the DC and AC out channels. These signals are straight from the PD in counts, before the filtering to undo the D1201111 pre-amp listed in 81868. PWM is at 5kHz, as can be seen in the spectrum.

TITLE: 01/14 Day Shift: 1530-0030 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Lock Acquisition
INCOMING OPERATOR: Ibrahim
SHIFT SUMMARY: Currently relocking, we just lost lock at LOWNOISE_ESD_ETMX.
LOG:                                                                                                                                                                           

During one of the large eqs over the weekend all of the BSC-ISI tripped. I checked one of the chambers and the trip was due to large low frequency drive railing the stage 2 horizontal actuators. The earthquake was big enough SEI_ENV went LARGE_EQ, well after the IFO lost lock, but also after the ISIs started tripping. In SEI_ENV I've reduced the threshold on the peak mon to 6000, about 1.5x the largest eq we've ever ridden out, down from 10000. I've also changed the stage 2 blends used in this state to some 1.5hz blends which roll-off the gs13s at low frequency much more aggressively. I don't think these changes would have been enough to prevent the ISI trips for this particular earthquake, the ISIs started tripping while peak mon was around 4000, so there must have been a large amount of ground motion that was out of band for that channel.

TITLE: 01/15 Eve Shift: 0030-0600 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Lock Acquisition
OUTGOING OPERATOR: Ryan C
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 2mph Gusts, 0mph 3min avg
    Primary useism: 0.03 μm/s
    Secondary useism: 0.39 μm/s
QUICK SUMMARY:

IFO is LOCKING at MOVE_SPOTS

The LN2 truck is yet to arrive and so it may cause a lockloss.

In 81638 we removed a DARM1 FM1 boost because of a glitch when ramping it off causing locklosses during ESD transitions in preparation to switch back to ETMX. Today the CDS team updated H1OMC0 models with an improved filter ramping: 82263. We hope this will allow us to keep the boost us which gives us more range against high microseism while relocking (it's always off by NLN).

Uncommented line 3058 from ISC_LOCK.py and reloaded: ezca.get_LIGOFilter('LSC-DARM1').turn_on('FM1'). If we have locklossses at ISC_LOCK state 557 or 558, at the operator can re-comment this line out. Tagging OpsInfo.

J. Kissel, T. Sanchez, E. Goetz, L. Dartez

*EDIT* This limitation is only pulling out data with test points from the A0/B0 filter outputs due to them being recorded in single precision, not double precision. The actual data for all internal calculations is double precisions, and in fact for the final calibrated gravitational wave strain is both calculated in, and then stored in frames as, double precision.

Back in July 2024 when I started to characterize the super-16 kHz-Nyquist frequency data off the OMC DCPDs with the live 524 kHz channels. See LHO:78516 for the whole story, but we got stalled when we ran into what we believed was some sort of single-precision, numerical precision noise, limited at the equivalent of 1e12 [A/rtHz] DCPD current or 1e-6 [V/rtHz] ADC voltage.

In LHO:78559, we ruled out single-point precision calculation of the ASD when we ran the same DCPD signal through a special version of DTT which uses double-precision to calculate the pwelch algorithm. That version of the data proved that the high frequency limitation is still there, and NOT the precision of DTT.
In that same data set, we also showed that if you ask DTT to remove the mean, i.e. large DC component of the signal, it also did NOT have any impact on this limit.

And it's in removing the mean that we reveal / confirm that it is "single-point precision" limit in the test point readbacks. 
Check out 1st attachment which is the data from LHO:78559, but with no DTT calibration applied. That means the channels are calibrated into the units of whatever they are coming off of the front-end -- in this case milliamps, or [mA].

The DC value of the test point channel during the time of measurement was ~20 [mA].

The front-end computes all its filtering in double precision, but the readbacks of the products of those calculations are single-point precision. An IEEE 754 32-bit base-2 floating-point, single precision channel has 24 bits of significance (excluding the sign and exponent) to hold the entire frequency dependent content of the time-series that has a DC value of ~20 [mA]. The (front end filtering algorithm?) rounds the 20.43 DC component to the nearest 2^n value, i.e. 2^5 = 32 ["mA"]. 

Eq. 2.2 of Liquid instruments article on quantization noise suggests that the amplitude spectral density of 1 bit spread across the 0 to 2^18 Hz (f_Nyquist) frequency range over which we care is 
    n_{ASD,RMS} = sqrt( DELTA^2 / (12 * f_Nyquist) ) 
                
                = DELTA * sqrt( 1 / 12 * 1/f_Nyquist )  
where 
    - DELTA is the minimum step resolution (i.e. the peak value / number of significant bits), 
    - the factor or 1/12 comes from the expectation value of the noise power derived from the integral of the product of instantaneous noise power and the probability that that power is distributed across one, specifically the least significant, bit
        (and we use the square root because we want the amplitude not the power)
    - the factor of 1/f_Nyquist comes from spreading out the (presumably frequency independent) power over the entire frequency range
        (and again, we use square root because we want the amplitude not the power)

In line-by-line math, that's 
     [[ 32 ["mA"_DC]              peak value
        * (1/2^24)                significant bits in single precision) ]]
     [[ 1 bit 
        * (1/sqrt(12))              expectation value quantization noise amplitude spread across 1 bit
        * (1/sqrt(2^18 Hz))         quantization noise spread across 0 to Nyquist frequency range
        ]] 

     = [ 32 / (2^24) ] * sqrt[ 1 / (12 * 2^18) ]

     = 1.07539868e-9 ["mA"/rtHz]
 
This number *exactly* the high-frequency asymptote we see. BINGO!

The next step was then to create pick-off paths of the ADC channels and high-pass -- i.e. remove the large DC component of the signal -- in the front-end. Importantly, this has to be the *first* filter, so the DC component is removed before any other calculation is done. We added the infrastructure and installed the filtering (LHO:78956, LHO:78975), but have not come back to the data until today.

Today, we're finally looking at the front-end OMC DCPD data that's been high-pass filtered with a 5th order, 1 Hz high-pass, with 40 dB stop-band attenuation and a 1 dB ripple 
    ellip("HighPass",5,1,40,1)
The odd (5th) order means that DC component is completely suppressed, leaving only the remaining frequency-dependent accumulated RMS, which is 5.8901e-4 [mA_RMS] to upper limit of the dynamic range and define the precision limit.

SIDE QUEST -- Fractional numbers are much less intuitive to "just round up to the nearest power of 2^n," so the equivalent of "converting 20.43 [mA] to 32 ["mA"]" for 5.8901e-4 is instead a process of,
    Converting 5.8901e-4 to floating point binary, 
        # sign exponent fraction
          0 01110100 00110100110110111110111
        # round up the fraction part
          0 01110100 01000000000000000000000
        # convert back to decimal
        0.00061035156
That takes the quantization limit down to 
     = [ 6.1035156e-4 / (2^24) ] * sqrt[ 1 / (12 * 2^18) ] 
     = 2.05e-14 ["mA"/rtHz]


Take a look at 2nd attachment, which compares the normal nominal OMC DCPD data to this 1 Hz high passed data. One can see all of the AA filtered data all the way out to the 232 kHz Nyquist frequency, because that data only goes as low as 1e-13 [mA].

Finally in the last attachment, we re-cast this into ADC noise units, to show where the data against the trace we'd had as a bench mark before. Note *this* comparison is a false comparison because the digital AA filtering is applied after the ADC noise is added. So -- don't read anything into the fact that the resolved noise goes below the ADC noise -- it's just a guide to the eye and a bench mark to remind folks that the numerical precision limit is *not* ADC noise.

In conclusion, if we want to investigate the OMC DCPD data above 7 kHz with test points, we need to make sure to use a version where the data is high-passed significantly.
So, now we can actually begin doing that...

FAMIS26026

Last week's report - alog82184. All spectra look good to me and agree with last week's report.

FAMIS31405

I replaced both sock filters for fresh ones and inspected the radiator air filters.

WP12274

FAMIS28946

We rebooted the h1guardian1 machine today for 3 things:

1. Point the machine back at nds0 as the primary nds server
   • I noticed the other day that the guardian was still defining the chosen nds server as nds1 primary and nds0 as secondary. I'm not entirely sure when this was changed, but maybe 2 years ago (alog66834).
   • This was done by changing the NDSSERVER definition in the /etc/guardian/local-env file
2. Relieve any stale processes that might latch the gps leap second data.
3. Quarterly machine reboot FAMIS task

All 168 nodes came back up and Erik confirmed that nds0 was seeing the traffic after the machine reboot.

Francisco, went to End X to do a beam movement, and I asked him kindly to just block a few beams for me, as a test of  the threshholds for PCALX_STAT Guardian.
Francisco blocked the beam for more than 60 seconds which should have reduced the TXPD values to their minimum values.

Conditions to be met:
CAL-PCALX_TX_PD_WATTS_OUTMON > 0.002

This minute of time where TXPD is below the minimum threshhold should have taken PCALX_STAT into A fault mode. 
But it never did.
This is because I changed the PCALX_STAT ISC_LOCK state prerequisite from 600 to 10 for this test today. But ISC_LOCK was in PREP_FOR_LOCKING[9] Instead of IDLE[10] which is what I expected to remain in for the duration of maintenance. In hind sight I should have just taken the ISC_Lock depenence out of the Guardian node completely.

Thanks goes out to Francisco for giving me a hand on a busy day. 

I noticed on the ops overview last night that there was an orange "Stand down alerts query failure" light. Looking at the journald log located on the ext-alerts vm, looks like it crashed on Jan 11, so our automated stand downs haven't been working since then, but the incoming events were unaffected. The error was that it wasn't able to connect to the superk site after so many retries. I've restarted this for now with the hope that it was just some maintenance or similar on thier end, but if it crashes again I'll find a more robust solution.

Tue Jan 14 10:03:25 2025 INFO: Fill completed in 3min 22secs

TCmins [-65C, -63C] OAT (0C, 32F)

Sheila, Mayank

We wanted to double check the calibration of IM4 trans into power on PRM, similar to 63812 and 62213

• 10 Watts, single bounce ITMY 16:24 UTC
• OMC locked 16:35 UTC.  Sensor correction was off so this alignment is wobbling more than it otherwise would. 
• sensor correction on 16:40 UTC - 16:45 UTC.  1420908058 for 300 seconds
• IM4 trans and AS_C whitening gains were 18dB at the time of this measurement. 
• Medians:
  • H1:IMC-IM4_TRANS_NSUM_OUT16 9.855005
  • H1:IMC-PWR_IN_OUT16 9.997126
  • H1:ASC-AS_C_SUM_OUT16 0.007815647
  • H1:OMC-DCPD_SUM_OUT16 15.558
• Using Craig's code here to get all the optic transmissions (he cites where the numbers came from at the top of the file), we expect 0.244% of the power incident on PRM to reach HAM6 in single bounce. 
• If we continue to trust the calibration of AS_C, and the known transmissions that Craig collected, we would apply a correction of 0.9566 to IM4 trans.

All h1omc0 models (h1iopomc0, h1omc, h1omcpi) were restarted with version 5.3.1 of the RCG.  This version changes the ramp function when turning on and off ramp-switched filters from a linear ramp to the so-called quadratic ramp. 

A linear ramp has the formula r = q  where q = (t - t0)/(t1 - t0), the new ramp has the formula r = -q⁴ + 2q², which has the nice property that dr/dq = 0 when q = 0 or q = 1.

This change is an attempt to address the problem described here: 81638

TITLE: 01/14 Eve Shift: 0030-0600 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Observing at 162Mpc
INCOMING OPERATOR: TJ
SHIFT SUMMARY:

IFO is in NLN and OBSERVING as of 04:47 UTC

Smooth shift with one Lockloss caused by the ETMX glitch (alog 82254)

Per Ryan C's instruction from Ryan S, I also adjusted the PSL ISS Ref signal (attached pic) to bring the refracted power above 4.

There was one SDF Diff that I accepted (also attached).

LOG:

None

Sheila, Jennie W

During commissioning window today we measured the P2L and Y2L gains for the PR2 mirror.

This is part of the work to understand how to move the beam on PR2 in order top avoid a reflection and subsquent stray beam off the PR2 scraper baffle (as found in alog #77631 by Annamaria and Robert).

The P2L gain was close to optimised and has been left at -0.36.

The commissioning window ran out before I could finish optimising the Y2L, I have left it at -7.00 which gave better A2L decpupling than the previous value.

Method:

• Inject line on PR2 M3 stage pitch at 30Hz using the OSC 9 filter bank in the ADS.
  • sitemap-> ASC -> Dither overview
• Check the PIT9 filter bank at the bottom of the page is not being used (CLK gain = 0).
• Set the matrix at the right of the row to have a 1 in the PIT9 to PR2 cell.
• Check there is a ramp time in the PIT 9 filter bank.
• Set the frequency to 30 and step the amnplitude to 1 in a couple of steps.
• Measure a reference for DARM and PRCL. The red and blue traces show the case when injecting into pitch with nominal A2L gains.
• Set the aevraging to exponential and zoom in on the line.
• Change gain in HzTS->PR2-> M3 DRIVEALIGN->P2L

 Do this until the line height relative to the background is minimised, allowing time for the ramp and a couple averages each step. One thing to note is that the line height above the background in the PRCL spectrum (LSC-PRCL_OUT_DQ) is much less than the relative height above the background in DARM which suggests that the noise in PRCL is not coupling strongly to DARM.

Switch off the line and set the PIT 9 amplitude to 0 counts.

Then repeat the above for the YAW 9 oscillator channel in the ADS, changing the gain in the Y2L gain filter bank in order to minimise the line height. The reference values for yaw are in the photo linked above with the green line showing DARM (we had gone into NLN cal meas with SRCL FF switched off for Elenna's SRCL tests so that is why DARM shows elevated noise). The orange line is the reference value of PRCL using the nominal Y2L gain. The same amplitude was used in the oscillator for this line but it was much higher above the PRCL background noise than for the pitch degree of freedom.

Template is at /ligo/hom,e/jennifer.wright/git/2025/A2L/20250113_A2L_PR2.xml