Reports 1-1 of 1 Clear search Modify search
VIS (SR2)
fabian.arellano - 15:58 Wednesday 25 May 2022 (20873) Print this report
Inter-calibration between LVDT, geophone, and IXV seismometer.

I did preparation to inter-calibrate the seismometer on the 2nd floor (IXV) with the LVDTs and geophones. The data will be acquired on a day with bad weather.

Concept

As vector quantities, the readout from the three types of devices are related according to this equation: GEO = SEIS + LVDT, where

  • GEO measures the displacement of the IP with respect to an inertial reference frame,
  • SEIS measures the displacement of the ground with respect to the inertial reference frame and
  • LVDT measures the displacement of the IP with respect to the ground.

Inter-calibration is finding a constant number a such that a * SEIS = GEO - LVDT, where the geophones and the LVDTs have been inter-calibrated already.

In practice, the subtraction of the time series can be done using the filter banks that are usually used for blending the LVDT and the geophones (alternative approches are explained in klog 20410).

Implementation

  • I added a gain of -1 to each of the filter banks SR2_IP_BLEND_{LVDTL,LVDTT}.
  • I added a gain of +1 to ech of the filter banks SR2_IP_BLEND_{ACCL,ACCT}.
  • Then, the subtraction GEO - LVDT in L and T are in the channels K1:VIS-SR2_IP_IDAMP_{L,T}_IN1, provided that the setpoints are set to zero (although this might not matter because diaggui subtracts the DC offset).

This implmentation shoud be used in TWR_FLOAT state, where the setpoint offsets can be turned off safely while keeping the suspension in the nominal position. I turned the subtraction filters on and updated the SDF in SAFE state. I'll get rid of these filters after I finish the measurements.

Images attached to this report
Comments to this report:
fabian.arellano - 16:29 Tuesday 27 September 2022 (22217) Print this report

Today I came back to the process of intercalibration of the seismometer on the ground (2nd floor) with the LVDT and the geophones. See entry 20873 for details.

I double-checked the intercalibation between the LVDTs and the geophones by measuring the IP-Y transfer function (see Fig. 1). The results seemed fine. Then I measured the seismometer output (in units of µm/rtHz) SEIS, and the substraction GEO - LVDT as described in entry 20873 (see Fig. 2).

I'll analyze the data and write comments in a future entry.

  • Directory: /kagra/Dropbox/Subsystems/VIS/vis_commissioning/sr2/inter_calibration/
  • File for the LVDT and geophone intercalibration assessment: IP_Y_EXC_displacement_cnt_new_geo.xml
  • File for the intercalibration of the seismometer on ground with the LVDT and geophone:  IP_Y_EXC_intercalibration_with_seis.xml
Images attached to this comment
fabian.arellano - 15:39 Wednesday 28 September 2022 (22229) Print this report

Using the data measured yesterday (klog 22217), I calculated the calibration factor between the output of seismometer on the ground (2nd floor) SEIS and the term (GEO-LVDT).

Per klog 20873, the aim is to calculate a such that a * SEIS = GEO-LVDT, in a frequency range where there is high coherence (Fig. 1). The value of a was calculated by minimizing the RMS value of the function  (GEO-LVDT) - a * SEIS.

The results are as follows:

  • Frequency range: from 160 to 500 mHz.
  • IP-L, Y Arm: a = 1.21104 (Fig. 2).
  • IP-T, X Arm: a =  1.17220 (Fig. 3).

These calibration factors still need to be written in the appropriate filter banks, and then the performance of the sensor correction can be assessed.

Images attached to this comment
fabian.arellano - 17:05 Thursday 29 September 2022 (22240) Print this report

Following what was done in Type-A for LVDT sensor correction using the seismometer (klog 21957), for SR2 I copied and (slightly) modified Terrence's filter for SRM.

The appropriate procedure for designing this filter is explained in Terrence's thesis in Sec. 8.3.3. Such a procedure takes into account the noise of the sensors, which is expected to be different for different suspensions. Terrence's filters were optimized for SRM, but we are using them in other suspensions as approximations.

Details of the modifications referred to above will follow in a future entry.

fabian.arellano - 11:49 Tuesday 18 October 2022 (22550) Print this report

Given the work reported in klogs 22541 and 22240, I calculated again the intercalibration factor between the seismometer on ground, and the LVDTs and geophones per klogs 22229 and 20873.

The results are as follows:

  • Frequency range: from 140 to 500 mHz (see coherence in Fig. 1).
  • IP-L, Y Arm: a = 1.196 (Fig. 2).
  • IP-T, X Arm: a =  1.175 (Fig. 3).

I added this values to the filter banks at SENSCORR.

This is information about the files I used:

  • Directory: /kagra/Dropbox/Subsystems/VIS/vis_commissioning/sr2/inter_calibration/
  • XML file: SR2_seis_intercalibration_with_LVDT_and_GEO.xml
  • Python script: seismometer_OnGround_Intercalibration.ipynb (it reads data directly from the xml file).
Images attached to this comment
Search Help
×

Warning

×