KAGRA Logbook

MIF (ITF Control)

masayuki.nakano - 22:11 Sunday 01 September 2019 (10246)

Weekend report

[Yokozawa, YamaT, nakano]

- SRFPMI

* Although the center part (SRMI) itself can be stably locked, we could not hold the SRMI with the arms close their resonance, even lock the ALS CARM with high bandwidth.
* Lock loss did not happen when the cavity pass very close the resonance, but happen little bit far from the resonance by about a few 100 Hz.Â
* I did not investigated so much, so I gonna look into it as the next step.

- Fine alignment
* Since I have found that the beam position on the ITMX, ITMY and SRM is not really good with the FPMI aligned state, I start a fine alignment of the IFO.

* GR QPD on the TMSX was aligned with the good cavity axis.
* DC ASC loop for the GR-X was implemented. The configuration of each matrix and servos is shown in the attached picture. I have not implemented into the guardian due to the DGS trouble. Also any measurements has been done such as a UGF, spectrum etc.

Images attached to this report

Comments to this report:

yutaro.enomoto - 9:42 Monday 02 September 2019 (10254)

> * Lock loss did not happen when the cavity pass very close the resonance, but happen little bit far from the resonance by about a few 100 Hz.

This makes me think that a f1 sideband of the 1st order HOM causes the issue when it goes across the resonance.
See 9505. Currently the f1 frequency is ~200Hz lower than the resonant frequency of the 1st order HOM:
f1 = 5.6244235e6 * 3 Hz = 35.010 kHz + 337* FSR
Transverse mode spacing = 34.82 kHz

So, INCREASING the frequency of the RF seed generation (~5.6 MHz) by 100 Hz or so would make some changes, I think.
You can increase more if you want, but f2 or 2*f2 might be dropping off from the top of the resonance of the IMC.

masayuki.nakano - 6:13 Tuesday 03 September 2019 (10276)

Yutaro,

Thanks for your comment, and yes, that is what I was thinking the first step to do. Actually this calculation helps a lot, thanks so much.
Also, I would like to try decreasing the bandwidth of the MICH and the SRCL loop. Comparing to the FPMI locking procedure, the MICH loop UGF is higher and it might feed back the fast garbage due to the arm resonance.

masayuki.nakano - 16:26 Tuesday 03 September 2019 (10286)

I changed the modulation frequency from 5.6244235 Mhz to 5.62443350 MHz.

masayuki.nakano - 7:32 Wednesday 04 September 2019 (10311)

I tried to lock SRFPMI but still it did not work.

Configuration I tried:
Yarm aligned, Xarm misaligned, MICH and SRCL aligned.
IR is locked to Yarm with ALS.

- Even I shifted the modulation frequency, the situation did not change so much.
- When the arm cavity gets close to the resonance, high frequency component (around 400 Hz) is appear in the 3f signals.
- Where the fast component start to appear might be comparable with where the lock loss happen.
- 400 Hz is comparable with the frequency difference between the 1st-order mode and the f1 frequency. So it might caused by that. I will compare the error signal with several modulation frequency and see the difference tomorrow.
- Also, it would be wothfull to put a notch filter for the SRCL, MICH servos.

Images attached to this comment