I took Jeff's measruements for calibrating the PRM OSEMs [LHO: 87102] to calibrate the M1, M2 and M3 stages. [See similar calcs for SRM here: LHO: 87166]. I only analyzed the data for the ALIGNED position of the suspension.
The TL;DR is that the M1 calibrations are in line with expectations from PR3/SR3/SRM.
The M2 calibrations are all over the place, with the Left side OSEMs seeing the most chenges (factors of 3.6 for LL and 2.2 for LR)
The M3 calibrations are really off for the Left side OSEMs (factors of 10 (!) for LL and 4 for LR).
My expectation is that the yaw alignment is really pushing these OSEMs to a point where the nonlinear response has kicked in enough that the ir response is all wonky. I think we need to analyze this information more before drawing any snap conclusions. I will do this on Monday to be ahead of Tuesday maintenance
Here is the full output of the calibration script [see attached figures for the before and after comparison]:
______________________________________________________________________
OSEM calibration of H1:SUS-PRM
Frequency range for calibration: 6 to 15 Hz
Stages to be calibrated: ['M1' 'M2' 'M3']
Measurement date: 2025-09-23_1810 (UTC).
%%%%%%%%%%%%
Stage: M1
%%%%%%%%%%%%
The suggested (calibrated) M1 OSEMINF gains are
(new T1) = 1.365 * (old T1) = 1.646
(new T2) = 1.265 * (old T2) = 1.370
(new T3) = 1.544 * (old T3) = 1.727
(new LF) = 1.702 * (old LF) = 1.840
(new RT) = 1.499 * (old RT) = 1.504
(new SD) = 1.627 * (old SD) = 1.808
To compensate for the OSEM gain changes, we estimate that the H1:SUS-PRM_M1_DAMP loops must be changed by factors of:
L gain = 0.627 * (old L gain)
T gain = 0.614 * (old T gain)
V gain = 0.726 * (old V gain)
R gain = 0.726 * (old R gain)
P gain = 0.719 * (old P gain)
Y gain = 0.627 * (old Y gain)
The calibration will change the apparent alignment of the suspension as seen by the M1 OSEMs
NOTE: The actual alignment of the suspension will NOT change as a result of the calibration process
The changes are computed as (osem2eul) * gain * pinv(osem2eul).
Using the alignments from 2025-09-23_1810 (UTC) as a reference, the new M1 apparent alignments are:
DOF Previous value New value Apparent change
---------------------------------------------------------------------------
L -78.6 um -125.4 um -46.8 um
T 83.9 um 136.5 um +52.6 um
V 50.6 um 67.5 um +16.8 um
R 64.4 urad 115.3 urad +51.0 urad
P -1229.3 urad -1507.6 urad -278.4 urad
Y -45.2 urad 27.0 urad +72.2 urad
%%%%%%%%%%%%
Stage: M2
%%%%%%%%%%%%
The suggested (calibrated) M2 OSEMINF gains are
(new UL) = 1.162 * (old UL) = 1.428
(new LL) = 3.622 * (old LL) = 4.249
(new UR) = 1.141 * (old UR) = 1.344
(new LR) = 2.281 * (old LR) = 2.589
The calibration will change the apparent alignment of the suspension as seen by the M2 OSEMs
NOTE: The actual alignment of the suspension will NOT change as a result of the calibration process
The changes are computed as (osem2eul) * gain * pinv(osem2eul).
Using the alignments from 2025-09-23_1810 (UTC) as a reference, the new M2 apparent alignments are:
DOF Previous value New value Apparent change
---------------------------------------------------------------------------
L 13.3 um 58.9 um +45.6 um
P -981.3 urad -2050.3 urad -1069.0 urad
Y 645.5 urad 905.7 urad +260.1 urad
%%%%%%%%%%%%
Stage: M3
%%%%%%%%%%%%
The suggested (calibrated) M3 OSEMINF gains are
(new UL) = 1.368 * (old UL) = 1.645
(new LL) = 10.906 * (old LL) = 13.240
(new UR) = 1.522 * (old UR) = 1.709
(new LR) = 4.051 * (old LR) = 4.756
The calibration will change the apparent alignment of the suspension as seen by the M3 OSEMs
NOTE: The actual alignment of the suspension will NOT change as a result of the calibration process
The changes are computed as (osem2eul) * gain * pinv(osem2eul).
Using the alignments from 2025-09-23_1810 (UTC) as a reference, the new M3 apparent alignments are:
DOF Previous value New value Apparent change
---------------------------------------------------------------------------
L 68.5 um 551.8 um +483.3 um
P -1519.6 urad -11711.4 urad -10191.7 urad
Y -344.6 urad -6603.7 urad -6259.1 urad
We have calculated a GS13 to OSEM calibration of H1 PRM ['M1' 'M2' 'M3'] using HAM2 ST1 drives from 2025-09-23_1810 (UTC).
We fit the response PRM_OSEMINF/HAM2_SUSPOINT to unity between 6 and 15 Hz to get a calibration such that 1 [OSEM m] = [GS13 m]
This message was generated automatically by OSEM_calibration_master.py on 2025-09-27 02:12:10.458163+00:00 UTC
%%%%%%%%%%%%%%%%%%%%%%%%%%%%
EXTRA INFORMATION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%
Stage: M1
%%%%%%%%%%%%
The H1:SUS-PRM_M1_OSEMINF gains at the time of measurement were:
(old) T1: 1.206
(old) T2: 1.083
(old) T3: 1.119
(old) LF: 1.081
(old) RT: 1.003
(old) SD: 1.111
The matrix to convert from the old PRM M1 Euler dofs to the (calibrated) new Euler dofs is:
+1.601 -0.0 +0.0 +0.0 -0.0 -0.008
+0.0 +1.627 -0.0 +0.0 +0.0 +0.0
+0.0 -0.0 +1.385 -0.001 +0.002 -0.0
-0.0 +0.0 -0.326 +1.385 -0.035 +0.0
+0.0 -0.0 +4.675 -0.28 +1.404 -0.0
-1.265 +0.0 -0.0 -0.0 +0.0 +1.601
The matrix is used as (M) * (old EUL dof) = (new EUL dof)
The dof ordering is ('L', 'T', 'V', 'R', 'P', 'Y')
%%%%%%%%%%%%
Stage: M2
%%%%%%%%%%%%
The H1:SUS-PRM_M2_OSEMINF gains at the time of measurement were:
(old) UL: 1.229
(old) LL: 1.173
(old) UR: 1.178
(old) LR: 1.135
To compensate for the M2 OSEM gain changes, any controllers using the M2 OSEMs as inputs must be compensated with gains of:
L gain = 0.611 * (old L gain)
P gain = 0.610 * (old P gain)
Y gain = 0.506 * (old Y gain)
The matrix to convert from the old PRM M2 Euler dofs to the (calibrated) new Euler dofs is:
+2.052 -0.043 -0.016
-18.856 +2.052 +0.33
-7.135 +0.33 +2.052
The matrix is used as (M) * (old EUL dof) = (new EUL dof)
The dof ordering is ('L', 'P', 'Y')
%%%%%%%%%%%%
Stage: M3
%%%%%%%%%%%%
The H1:SUS-PRM_M3_OSEMINF gains at the time of measurement were:
(old) UL: 1.202
(old) LL: 1.214
(old) UR: 1.123
(old) LR: 1.174
To compensate for the M3 OSEM gain changes, any controllers using the M3 OSEMs as inputs must be compensated with gains of:
L gain = 0.424 * (old L gain)
P gain = 0.431 * (old P gain)
Y gain = 0.272 * (old Y gain)
The matrix to convert from the old PRM M3 Euler dofs to the (calibrated) new Euler dofs is:
+4.462 -0.144 -0.08
-63.208 +4.462 +1.752
-35.107 +1.752 +4.462
The matrix is used as (M) * (old EUL dof) = (new EUL dof)
The dof ordering is ('L', 'P', 'Y')
