KAGRA Logbook

CAL (General)

takahiro.yamamoto - 1:04 Thursday 06 October 2022 (22366)

DARM calibration was updated

Abstract

Open loop transfer function of DARM was measured with more wide frequency ranges.
Our transfer function model agrees well with measurements as shown in Fig.1 and Fig.2.

Details

Current DARM model already contains also all circuits response that zpk parameter is theoretical values not measured value.
According to the latest measurement, DARM optical gain from displacement to DARM_IN1_DQ is ~8.67dB (Old value was 8.45dB. So it's a change as ~3%.)
Calibration was updated around 19:48 based on this measurement and now h(t) is provided with new calibration parameters.

Now rough calibration which is subtracting all effects excepting optical gain from measured transfer function can be done from the commissioning dock.
It works with only a few clicks if there is no change in DARM control configuration e.g. change of engaged FMs on DARM1, DARM2, LOCK_L, etc.,
but checking the existence of change in DARM control configuration is the most hard work...
klog#21733 may be slightly helpful to check it.

Images attached to this report

Comments to this report:

takahiro.yamamoto - 2:54 Friday 07 October 2022 (22396)

I found one or two mistakes in the polarity of the calibration factors.

One is just my mistake to treat DARM2 gain as 1.0 in DARM model. It is -1.0 in correctly.

Another one is about relation between +L of DARM and +L of VIS.
I will check it more carefully tomorrow.

takahiro.yamamoto - 12:32 Sunday 09 October 2022 (22419)

[Ushiba, YamaT]

This is Friday's activity.

Abstract

- Mistakes in DARM calibration was fixed and applied from 19:18 on Oct. 7th.
- DARM rough calibration was automated as shown in Fig.1.

Details

Calibration fix
We checked polarities of all calibration parameters and confirmed that old calibration was wrong as I guessed in klog#22396.

Current calibration is based on klog#21089.
"Actuator efficiency" means a conversion factor from count at SUMOUT_L_OUT to the displacement.
After this measurement, TM_COILOUTF_{H1,H2,H3,H4}_GAIN were changed as two times larger (from normalized by 0.5 to normalized by 1.0) reported as klog#22314.
In these reports, there is no polarity information but the polarity of COILOUTF_GAIN for all suspensions are set as when SUMOUT_OUT is positive, suspension moves +L direction in suspension local basis (It's very helpful definition. Why we couldn't do so in O3GK...)
And also, +L direction of ETMX is -L for DARM because DARM = (Lx - Ly)/2.
So we need to set actuator efficiency of ETMX as a negative value.

Automation of rough calibration
Roughly estimation of DARM optical gain was automated as shown in Fig.1.
Please choose diaggui file of DARM OLTF measurement.
This script pick up all foton filter designs, and EPICS values of filter switches, filter gains, matrix values etc. at the measurement time which is picked up from the diaggui xml file.
(In current implementation, a cavity pole frequency is fixed and only optical gain is a free parameter.)
Latest results are shown as Fig.2 and Fig.3.

Images attached to this comment