Betsy, Rahul
Last week Betsy and I replaced the bottom wire of FC1 which was found to be of incorrect thickness (see alog 60570, 60513 and 60532) and we were missing 3.21Hz peak for P DoF when we took TF as that time.
After replacing the wire with correct thickness (0.005 inch diameter, verified using calipers) I spent some time correcting the pitch on IM and bottom mass by transferring 10gm from the back of the IM to the front (took some iterations to get it right). Next, I took all six DoF transfer function measurements and the results are attached below.
I can confirm that the suspension looks healthy and is free of rubbing - all the peaks are now visible for all 6 DoF (including the 3.21Hz Pitch DoF). The magnitude for some of the DoF shows an offset - especially for L Dof and T DoF, however I believe this is an electronics issue rather than mechanical. I will discuss this with Jeff Kissel and sort it out.
Note - I found that the excitation amplitude for almost all 6 DoF were too high and were causing the SUS to saturate, I had to reduced them by 50% to be able to run the TF. At the same time I made sure that the coherence didn't drop below 0.9.
The templates and text files are located at the following location, I will post the matlab plots later on.
/ligo/svncommon/SusSVN/sus/trunk/HSTS/H1/FC1/SAGM1/Data
2021-11-23_2000_H1SUSFC1_M1_WhiteNoise_L_0p02to50Hz.xml
2021-11-23_2000_H1SUSFC1_M1_WhiteNoise_P_0p02to50Hz.xml
2021-11-23_2000_H1SUSFC1_M1_WhiteNoise_R_0p02to50Hz.xml
2021-11-23_2000_H1SUSFC1_M1_WhiteNoise_T_0p02to50Hz.xml
2021-11-23_2000_H1SUSFC1_M1_WhiteNoise_V_0p02to50Hz.xml
2021-11-23_2000_H1SUSFC1_M1_WhiteNoise_Y_0p02to50Hz.xml
2021-11-23_2100_H1SUSFC1_M1_WhiteNoise_T_0p02to50Hz_tf.txt
2021-11-23_2100_H1SUSFC1_M1_WhiteNoise_L_0p02to50Hz_tf.txt
2021-11-23_2100_H1SUSFC1_M1_WhiteNoise_P_0p02to50Hz_tf.txt
2021-11-23_2100_H1SUSFC1_M1_WhiteNoise_R_0p02to50Hz_tf.txt
2021-11-23_2100_H1SUSFC1_M1_WhiteNoise_V_0p02to50Hz_tf.txt
2021-11-23_2100_H1SUSFC1_M1_WhiteNoise_Y_0p02to50Hz_tf.txt
Good to see this!
J. Kissel
I've processed the data in this aLOG in the usual fashion. Upon looking at comparisons with other suspensions, I got worried that the pitch resonances -- while there are now the two distinct peaks that we hope for with the corrected wire thickness -- in general are bit high.
So -- I embarked on a loft goal of finding all the latest in-vacuum measurements of HSTS, in hopes that I could survey the distribution of pitch frequencies -- in case this is just a feature of the design (say, like how the QUAD's reaction chain pitch mode frequencies are arbitrary, and determined by how ever the lacing cables are placed / distributed, etc.)
I only got so far before I had to give up, since we were much further behind on processing data than I'd hoped; see updates for the following SUS in the correspondingly indicated aLOG:
- H1SUSMC1: LHO:60757 and LHO:60758
- H1SUSMC2: LHO:60760
- H1SUSMC3: LHO:60763 and LHO:60764
- H1SUSSR2: LHO:60755
But anyways, I can at least share what I have -- see attached .pdfs for the complete set of data, but I focus in on the comparison of P to P transfer functions with
2021-11-23_H1SUSFC1_Phase3a_UnDamped_M1_P-P_TF_ZOOMED.pdf.
We see that the current response (2021-11-23, red) is indeed restored back to it's wire-build response (2021-01-29, cyan) after the M2 to M3 wire fix Rahul mentions above, *but* the highest frequency pitch modes are still, in general are still significantly higher than, say H1 PR2 (hence me trying to embark on the survey of these modes).
Interestingly, zooming out a bit, in page 5 of 2021-11-23_H1SUSFC1_Phase3a_UnDamped_M1_ALL_ZOOMED_TFs.pdf, you see that the first pitch mode is now much like other suspensions at 0.96 Hz (again, red trace), where it was quite high before when the wire thickness was wrong (again, green trace).
I this is a mystery worth chasing down before we move on:
Action Items:
- continue make sure we have processed all the latest and greatest data from all HSTS, then make plots comparing the pitch to pitch transfer functions in detail.
- noddle around with the matlab model, to see if there's another parameter that serves a smoking gun for this symptom.
If we find other SUS that have their highest pitch resonances this high up, then we can probably move on -- but at least from what I'm seeing of MC1, MC2, and SR2 -- they're (yes, different from the model, but) pretty consistent.
I'll pick up the study next week.
