KAGRA Logbook

MIF (ITF Control)

masayuki.nakano - 2:57 Tuesday 04 February 2020 (12763)

DC readout

[Enomoto, Nagano, Hasegawa, Yamamoto, Nakano]

We succeeded in locking the PRFPMI with DC-readout. The NS-NS binary range is ~40 kpc, as shown in the attached picture.

From now on, we can concentrate on noise identification and hunting, and I would say it would be possible to reach up to 1Mpc in two weeks.

-- Details --

The lock duration was about 1 hour. I kicked the IFO for measurement, so the lock was stable enough to continue to be locked for several hours.
The guardian can hand-off the TRAS signal to the OMC DC signal, although the OMC cannot be locked automatically. So the LSC LOCK guardian waits for the OMC to be locked after putting the DARM offset. I took over the OMC operation to Yokozawa-san and Yamamoto-sensei.
The noise limiting the sensitivity will be reported by Yamamoto-sensei.
DC offset is 30 pm, corresponds to 20 mW (2000cnts at OMC DC PD A). I put -0.5 cnts to TRAS servo filter.
OMC UGF is ~100 Hz. The gain of FB_FLT is 0.5 and FM2, FM6 was engaged. Once the OMC is locked, FM5 was engaged.
Dr. Okutomi made a beautiful IFO monitor screen as shown in the attached picture. We can easily find the BNS range. I really like it.

BTW

The OMC DC PD B has a noise around 1 kHz, which is not seen in PD A. Is this known?

Images attached to this report

Comments to this report:

takaaki.yokozawa - 3:05 Tuesday 04 February 2020 (12764)

Thank you for giving us the beautiful sensitivity curve!

I evaluated the noise floor sources. See the attached pdf file

As I reported last weekend, ETMY suspension may limit the noise.

Non-image files attached to this comment

bruco200204.pdf

takahiro.yamamoto - 3:36 Tuesday 04 February 2020 (12765)

DARM calibration.
- Calibration parameters were updated as latest (PRFPMI with OMC DC) ones.
- Measured OLTF by Nakano-kun is shown in fig1.
- Model TF is roughly consistent with measurement (see fig2).
- Optical gain (from Displacement to LSC-OMC_DC_IN1) is ~1.6e14 [ct/m]
- Calibration is roughly checked by the fact that (1/C)*v_err and A*v_fb cross over around UGF (see fig3).

Noise estimation
- Around 100Hz: MICH and/or PRCL limit DARM sensitivity (see top left panel in fig4).
- Above 1kHz: There is weak coherence with frequency and intensity noise. (see fig4).
(In high frequency, sensitivity is already very close to OMCPD dark noise.)
- After Nakano-kun applied roll-off filter for MICH, coherence between MICH and DARM around 100Hz was decrease. (see fig5.)
(But coherence around 30Hz (with MICH) and around 300Hz (with PRCL) were increase.

IFO guardian update
- The definition of LOCKED state was changed as 400(TRANSIT_TOOMCDC) <= K1:LSC-LOCK_STATE_N <= 1000(OBSERVATION)

Images attached to this comment

tomohiro.yamada - 4:51 Tuesday 04 February 2020 (12766)

Sensitivity improvement was clearly seen around 20-30 Hz by reducing gains of damping filters (EX TML, EY TML).

Although roll-off should be worked, we need to disable TM damping at some point.

tomotada.akutsu - 5:28 Tuesday 04 February 2020 (12767)

Great!!! Thanks so much for your continous, patient, and great efforts.

tomotada.akutsu - 5:39 Tuesday 04 February 2020 (12768)

About the coherence found with the TMSX-IR PD: did you also include some in-cavity-PDs for your bruco analysis? I mean the four PDs per a NAB. It might be possible to get clues by checking with them, if you have not done that.

takaaki.yokozawa - 5:44 Tuesday 04 February 2020 (12769)

Akutsu-san,

Sure, I calculated the coherence with all PDs. I will start the investigation soon.

Also, you can download the bruco result by typing

$ scp -r controls@k1ctr1:/users/DET/Result/Bruco/medm/CAL-CS_PROC_DARM_DISPLACEMENT_DQ-ALS-LSC-CAL/1264780218-300 ./

tomotada.akutsu - 5:52 Tuesday 04 February 2020 (12770)

Thanks for the information,,, but I started to suspect whether they have been turned on...

yutaro.enomoto - 15:25 Tuesday 04 February 2020 (12772)

= Note =
The optical power on OMC TRANS PDA was 3.4mW (not ~20mW)

= OMC PD dark noise and shot noise =

Here shows the expected contributions of PD dark noise and shot noise on DARM.
This indicates that there is another noise source in high frequency noise floor.

Images attached to this comment