Reports 1-1 of 1 Clear search Modify search
Detector Characterisation (Broadband noise)
mwas - 9:08 Thursday 29 November 2018 (43798) Print this report
Bump at 72Hz in hrec and B4_DC
Figure 1, there is a wide bump centered at ~72Hz and about 10Hz wide that is spoiling the sensitivity.
In purple is yesterday when we reach 40Mpc, and in blue Sunday morning with the larger dark fringe offset.
In both cases the bump is there, but yesterday it was higher by a factor 2 in B4 DC, and more clearly visible in h(t).

Figure 2, shows that the bump has been fluctuating in amplitude over the past 7 days in B4 DC.

The bump is not new, I could find it also in October data.
Figure 3, shows that in August there was also a bump in B4 DC, but it was a bit more narrow and jumping in frequency a little.

This looks similar to excitation caused by the step-motors in the minitower suspensions, but Alessandro has already checked that for SPRB it is switched off.

Figure 4, the bump in B4 DC, seems to be caused by a mechanical excitation at ~73.5Hz, that is visible in BS_ACC_Z, B4_GHOST_ACC_Z and PR_ACC_Z, and was much higher yesterday than on Sunday.
Images attached to this report
Comments to this report:
fiori - 17:11 Thursday 29 November 2018 (43810) Print this report

The noise from the source seems increased suddenly on Wed. night (Nov 28) around 3 UTC: see spectrogram1 and zoom-in.

around midnight of the 28th it did a peculiar up and down frequency pattern, which might be interesting for the line tracker tool.

the noise is also seen by the hall microphone (which also sees harmonics), looking at the amplitude in accelerometers the source should be located in the region between IB and BS.

Images attached to this comment
fiori - 15:56 Friday 30 November 2018 (43828) Print this report

the bump (which now moved to 71.Hz) seems still there after yesterday intervention on SPRB suspension: Figure 1 compares the lock before and the one after the intervention.

the bump looks a bit reduced in DARM, but larger on B4_DC, while the seismic vibration measured by accelerometers did not change significantly. One tentative interpretation is that the back scattered beam now impinges on a sligtly different location on the vacuum walls, but the coupling to DARM reduced slightly.

Additionally it now seems to disentangle another bump in DARM around 142Hz (double of 71Hz ...) which also corresponds to a bump in B4_DC.

Figure 2 instead compares 2:30 UTC tonight with Nov. 28, it is evident that the 70 Hz bump moves with the seismic excitation line measured by accelerometers on vacuum tanks and CEB hall microphone.

Images attached to this comment
fiori, paoletti, boschi - 15:12 Tuesday 04 December 2018 (43899) Print this report

Yesterday we investigated  the source of the vibration at the origin of the Hrec peak at 70Hz. We found a strong acoustic emission near the PR  SUSP rack.

We then placed one accelerometer (ENV_SR_ACC_Z) on top of the Sa_PR crate and we increased the speed of the fans upper layer (from 5 to 15).

We observed the line moving in conicidence with the speed increase action (Figure 1). See that the same line moves also in the accelerometer on PR vacuum tank (ENV_PR_ACC_Z) and it corresponds to our best knowledge to the line that we associated to the bump (see previous comments).

Unfortunately the interferometer was not locked, so we need to repeat the test in LN3 (ENV_SR_ACC accelerometer is kept there).

Images attached to this comment
fiori, direnzo - 13:23 Wednesday 05 December 2018 (43908) Print this report

During tonight lock the bump at 71Hz steadily increased in amplitude: see this Figure. As well also bumps at 115Hz and 140hz increases in a similar way: Figure. This seems to drive the drop of the range: Figure

All three bumps seem to have an associated bump in LSC_B4: Figure. However looks that the amplitude of these bumps in LSC_B4 was quite constant during the lock, and the same is for the amplitude of the seismic vibration of the rack fan (ENV_SR_ACC accelerometer is still sitting on Sa_PR crate).

It is interesting to check if the noise increase in DARM correlates to some misalignment which might have caused a change of  backscattering to DARM.

Images attached to this comment
ruggi - 19:34 Wednesday 05 December 2018 (43915) Print this report

A relevant drift is present on some signal coming from SPRB: B4 horizontal position on the camera and horizontal correction of B4_QD GALVO (fig 1).

Trying to find an evidence of ITF misalignment, which could explain the behavior of SPRB signals, one possible out of loop monitor is the dithering signal which measure the position of the beam on WI mirror (fig 2). A drift of about 100 um is present, but I don't know if it is enough to produce the effect on B4 camera. For a beam displacement like that a rotation of IN/END mirrors by about 0.1 urad is needed, and it looks a drift quite small. This behavior is confirmed by the payload optical levers (fig 3 - by the way, it is quite surprising to see those out of loop signals so stable: generally we expect that the reference would be lost by a couple of urad after several hours).

Also on SDB2 the beam looks stable enough (fig 4), and no particular adjustment is requested to SDB1 control to keep a good alignment (fig 5).

It would be interesting to do some experiment for a better understanding of the drift visible on B4.

Images attached to this comment
fiori - 14:36 Thursday 06 December 2018 (43927) Print this report

The first plot is a look at SPRB  position drift overnight and this morning.The sign of SPRB Pos X drift changed around 23 UTC, again around 01 UTC and again this morning (I do not know if this is a consequence of a voluntary action or else).

What seems interesting is that the DARM rms noise around 70Hz and 140Hz follows the sign of this change.

The second plot shows also the GALVO corrections of SPRB and also SDB2.

Images attached to this comment
fiori, paoletti - 10:30 Friday 07 December 2018 (43941) Print this report

the bumps moved and apparently disappeared yesterday around 20:22 UTC following a DECREASE in frequency of the vibration noise of the Sa_PR DSP crate's fan tray. I guess this was an intentional test (Valerio Boschi will comment).

Images attached to this comment
Boschi - 11:07 Friday 07 December 2018 (43943) Print this report

The speed level of both fan units of Sa_PR has been set to maximum (from about 2300 RPM to 2800 RPM) in order to test the effects on the sensitivity.

Search Help
×

Warning

×