The following AI Chassis were updated:

AI Chassis D1000305 S1108084 (SUSEY-C1, slot U32)
AI Chassis D1000305 S1108070 (SUSEY-C1, slot U31)
AI Chassis D1500177 S1500300 (SUSEY-C1, slot U26)

The rear panel and DAC AI Interface Board D1000551 were removed. A new D2400308 LIGO DAC Anti Image Chassis Rear Panel and LIGO DAC AI Interface D2500097 were installed.

"Yes, and..." to Fil's inventory and comments about what's changed within them --

AI Chassis D1000305 S1108084 (SUSEY-C1, slot U32) has now been transformed into a D2500353 AI chassis assembly
AI Chassis D1000305 S1108070 (SUSEY-C1, slot U31) has now been transformed into a D2500353 AI chassis assembly
AI Chassis D1500177 S1500300 (SUSEY-C1, slot U26) has now been transformed into a D2500400 AI chassis assembly

Richard has asked for a cell phone alarm on FCES VEA temperature. I have added the following alarm:

<Channel name="H0:FMC-FCES_VEA_SPACE_TEMP_DEGF" low="39" high="90" description="FCES VEA Temp">
 

which texts/emails Richard and Tyler if alarm persists for 10 minutes or more.

On a positive note the drop on temperature aides the internal pressure of all vacuum envelopes, internal to HAM8 and HAM8 annulus.  See attached plot for internal pressure of HAM8 for a change of about 2.0x10^-08 Torr.

The goal of these measurements is to use the ADF to measure the rotation of the squeezed state around higher order mode arm cavity resonances as described in section IV.B of LIGO-P2500132. Ultimately we would like to be able to use such measurements to diagnose the thermal state as a guide in how to tune TCS in order to improve mode matching.

While we are still digesting the results, the first plot shows the preliminary inferred squeezed state rotation for the three good data sets that we got. pre-step 1 and pre-step 2 are the two taken on Nov 22. (Everything turned out to be fine with the first sweep.) post-step is the last one taken this morning after the CO2s had been stepped and the IFO had been up for 7 hours. There is a small arbitrary constant offset in the SQZ angle of order a degree, so all angles have been shifted to be zero at 10.450 kHz. The second plot shows the OMC DCPD spectra 5 min before the start of each of these three sweeps.

Two rotation peaks are visible rather than the one shown in Fig 7 of P2500132. This is expected since the arms are astigmatic in reality, and our more detailed models show the same behavior. Each eigenmode is resonant at a slightly different frequency and is responsible for a rotation as in Fig 7 which are superimposed as observed in this data.

The post-step rotation is slightly smaller in magnitude and is shifted to slightly lower frequencies. Since the CO2 is mainly a higher order actuator, this is consistent with our expectations that this rotation is predominantly sensitive to quadratic mismatch with higher order aberrations altering the detailed behavior. However, seeing as how the pre-step was not taken just prior to the CO2 step in the same lock, I think it's just as likely that this is due to it being in a slightly different thermal state before the CO2 step. The peaks in the OMC DCPD spectra also appear slightly lower in frequency for the post-step.

I think the main takeaway from these measurements so far is showing that we can measure this rotation and resolve changes on the scale that our modeling suggests would be useful.

This analysis was done in the aligoNB environment by running
pytest /ligo/gitcommon/squeezing/sqzutils/analysis/T_10kHz_ADF.py --tb short -s -k T_Nov_CO2_step

I have updated the "o4" program for end of run.

              O4 started: 2023-05-24 08:00:00.000000 PDT [1368975618]
              O4   ended: 2025-11-18 08:00:00.000000 PST [1447516818]
                     now: 2025-11-18 09:36:41.000000 PST [1447522619]

               O4A_START: 2023-05-24 08:00:00.000000 PDT [1368975618]
                 O4A_END: 2024-01-16 08:00:00.000000 PST [1389456018]
               O4B_START: 2024-04-10 08:00:00.000000 PDT [1396796418]
          O4B_VENT_START: 2024-07-17 11:52:00.000000 PDT [1405277538]
            O4B_VENT_END: 2024-08-12 12:37:00.000000 PDT [1407526638]
                 O4B_END: 2025-04-01 08:00:00.000000 PDT [1427554818]
            O4C_STARTGPS: 2025-06-11 08:00:00.000000 PDT [1433689218]
                  O4_END: 2025-11-18 08:00:00.000000 PST [1447516818]
 

DC0 is down now as part of the DAQ 0-leg upgrade. Its GPS EPICS channels was being used by many systems on the CDS Overview MEDM screen, resulting in a lot of purple boxes. I've switch these over to DC1 and restarted the nuc20 FOM. If you have purple boxes on your CDS Overviews, please restart them.

VACSTAT is currently in the "NOT OK" condition due to missing DC0. I'll upgrade it later when Jonathan gets identical frames on the frame writers.

The dates between Jan 25th 2025 and April 1st 2025 are "O4c part 1".  

As shown on the JRPC wiki, which is meant to be our 'single source of truth' for run dates (although it's not especially well organized), O4b ended and O4c began during maintenance Tuesday on Jan 25th 2025.  O4c.Part1 (the "part 1" is an unofficial name) ran until the vent break on April 1st 2025.  O4c.Part2 (again, "part 2" is unofficial) ran from June 11th 2025 - Nov 18th 2025. 

I think Jenne meant to write Tuesday January 28th instead of the 25th which is a Saturday.

I reran these plots as the axis limits of the P_REFL vs. P_DCPD plot were wrong and were cutting off end point, plus removed line that explained what the y-intercept was for each plot as different dependent on whether we plot P_REFL vs P_AS or P_REFL vs. P_DCPD.

I also switched the P_REFL channel being used in my code to H1:OMC-REFL_A_DC_POWER from H1:OMC-REFL_A_DC_POWERMON as this latter channel has a factor of 100 relative to the former that I don't really undestand. The DC_POWER channel seems to give a realistic reading for the OMC reflected power that is much less than the power into HAM6.

First link contains two plots when we were half-way thermalised:

First plot is power at the OMC reflected PD vs. power at the antisymmetric port (calibrated into the power into HAM6) for the case where we were 1Hr 25 mins into lock.

Second plot is power at the OMC reflected PD vs. power transmitted to the DCPDs for the case where we were 1Hr 25 mins into lock.

The second link contains two plots also when we were thermalised:

First plot is power at the OMC reflected PD vs. power at the antisymmetric port (calibrated into the power into HAM6) for the case where we were 3 hrs 59 mins into lock.

Second plot is power at the OMC reflected PD vs. power transmitted to the DCPDs for the case where we were 2 hrs 59 mins into lock.

We can write down an equation for P_REFL from the OMC in terms of P_DCPD, using our linear fit.

P_REFL = ( c*P_DCPD + d ) mW

We know the nominal setting for P_DCPD is 40mA, and we can take the responsivity of the DCPDs at 100 % efficiency to be 

responsivity = e * lambda / (c * h)

From the squeezer budget (E2400269) we can get a number for the Q.E. of the PD that also includes OMC losses for the transmitted beam, transmissivity = 0.937

Therefore the power at the DCPDs at nominal DARM offset is 40mA/(responsivity*transmissivity) = 49.7 mW.

The reflectivity of the OMC breadboard is 2.75e-4 as calculated from parameters given in T1500060-v3.

The formula for the power measured at the reflected port of the OMC can also be expressed as:

P_REFL = R_cav [ P_00_arm + P_00_cd] + P_HOM_arm + P_sb + P_HOM_cd

where P_00_arm is the power in the fundamental carrier mode which changes with DARM offset, P_00_cd is the contrast defect light that is in the fundamental carrier mode, P_HOM_arm is light at higher order mode frequencies which also change with DARM offset, P_sb is power in the sidebands, P_HOM_cd is contrast defect light at higher order mode frequencies.

Whereas the formula for the power transmitted by the OMC can be described as:

P_DCPD = T_cav [ P_00_arm + P_00_cd]

As we assume that all HOM and sidebands are reflected by the OMC.

The mode-matching, MM, of the OMC to the differential arm mode is defined as:

P_00_arm / ( P_HOM_arm + P_00_arm )

This can be calculated using:

P_00_arm = (P_DCPD - d)/T_cav

P_HOM_arm = P_REFL - d - (R_cav*P_00_arm)

If we do this calculation for the half-way thermalised measurement we get 

MM = 0.9975

For the fully thermalised case we get:

MM = 0.9978

The code to calculate this is in: Calculate_refl_power.py located at /ligo/home/jennifer.wright/git/2025/DARM_OFFSET/

Correction to above calculation where I made a mistake in how to calculate P_00_arm:

The mode-matching, MM, of the OMC to the differential arm mode is defined as:

P_00_arm / ( P_HOM_arm + P_00_arm )

The power from higher order modes of the carrier from the differential arm motion can be calculated using:

P_HOM_arm = P_REFL - d - (R_cav*P_00_arm)

In order to calculate P_00_arm we need to use the measurement of contrast defect which can be obtained from our Sep 4th DARM offset measurements by running

python /ligo/gicommon/labutils/darm_offset_step/plot_darm_optical_gain_vs_dcpd_sum.py

on the offset step data we took on Sep 4th (see LHO alog #86744). This code plots the P_AS vs. P_DCPD graphs I mentioned in that alog, but also works our how the optical gain changes from the PCAL line height changes in DARM as the offset is changed. The plot of P_DCPD vs. optical gain has a minimum where the optical gain is zero and the power level here is considered the contrast defect (P_00_cd * T_cav) in our notation.

The contrast defect for the 255 Hz line is 1.202 mW for the partially thermalised interferometer and 1.268 mW for the fully thermalised interferometer.

Then we can obtain the power in the carrier from differential arm motion as:

P_00_arm = (P_DCPD - contrast defect)/T_cav

If we do this calculation for the half-way thermalised measurement we get 

MM = 0.9978

For the fully thermalised case we get:

MM = 0.9981

The code to calculate this: Calculate_refl_power.py located at /ligo/home/jennifer.wright/git/2025/DARM_OFFSET/ has been updated.

I plotted the loss between the antisymmetric port (calibrated into the power entering HAM6) to the power on the DCPDs. This is the inverse of the slopes in the graphs above.

All three are poltted on one graph, using plot_AS_vs_DCPD_changes.py in my own cope of the labutils repository at /ligo/home/jennifer.wright/git/local_git_copies/labutils/darm_offset_step/ .

Sheila and Camilla both agreed the loss for the two bottom lines (purple and red) are too high. These imply that a hot OM2 gives us over 20 % output losses.

If we look at the increase in loss from cold OM2 to hot OM2 this is a factor of 2.1 (210 % increase).

Compared to the decrease in optical gain squared (which we expect to reflect the change in output losses, which was:

(0.986^2 - 0.994 ^2) / 0.994^2 = -0.016 (1.6 % decrease).

We might have to check the alignment of out optics was not changing while we changed the darm offset.

Looking at OM1, OM2 and SRM alignment it did change during the darm offset steps with the biggest change (in the third offsset step measurement) being in OM2 pitch and yaw, this is only a change around 6 microradians (Elenna and Jeff state this calibration in correct to within an order of magnitude). Not sure if this enough to invalidate the loss values we measure. OM3 and OMC sus did not change much but this is because IU purposely unlocked the OMC ASC while changing the darm offset.

Jennie W, Matt T,

I plotted the antisymmetric power during the darm offset step vs. the power reflected by the OMC and am now very confused as the AS power looks to be smaller than the power reflected form the OMC. See the ndscope where I have zoomed in on the same time segment for both channels. The OMC-REFL channel is mean to be calibrated into mW and the ASC-AS_C channel is meant to be calibrated into W entering HAM 6 (even though the actual pick-off is the transmission through OM1).

The two plots attached show how the ratio between AS and OMC-REFL power changes during one of the DARM offset measurements we did right after I took this ndscope data.

Plot 1 hr 25 mins into lock.

Plot 2 hrs 59 mins into lock.

For each point the code returns the median of the time series at each step, this mioght be less valie for OMC-REFL as it is a lot noisier than ASC-AS_C.

I am still confused about the hogher power at OMC-REFL and wondering if:

a) I am confused about the calibration of one of these channels.

b) the calibration of one of these channels is wrong.

I plotted the three measurements of P_AS vs. P_DCPD during thermalisation on the same plot to make them easier to compare.

The code for this is in /ligo/home/jennifer.wright/git/local_git_copies/labutils/darm_offset_step/plot_OMC_REFL_changes/P_as_vs_P_DCPD__comp_Sep_04.pdf