stefan.ballmer@LIGO.ORG - posted 22:28, Wednesday 29 June 2016 - last comment - 13:05, Thursday 30 June 2016(28078)
Short lock
Sheila, Carl, Stefen
With the ISS 2nd loop patch (see alog 28076) we successfully reached the "lowish" nosie state again. The spectrum loked the same as in alog 27990.
.
Start: Jun 30 2016 05:01:36 UTC
Loss:Jun 30 2016 05:05:22 UTC
The loss was presumably cause by a 18.04kHz mode ringing up rapidly. Carl will add more details on that.
The spectrum at lockloss shows a line in OMC DCPDs and Arm transmission QPDs - see first image.
This is not the first time I have seen this line. On 22nd June I saw it but I thought it was asociated with Terra's mode excitation experiments. On the 22nd it appeared out of nowhere, with a time constant of 16 seconds. See second image.
Today the time constant was not as short, about 20 seconds. See the third image of the ringup and fit. Today I did not see any associated increase int he ITM drive signals.
We had 2 more of these short locks, the first we again got to "lowish" noise and lost lock sudenly after a few minutes, the next time we sat at increase power (40 Watts, none of our low noise steps) and lost it after about 10 minutes.
We reverted a ring heater change from a few days ago.
During the second short lock we started the calibration measurement, but had to abort when we lost lock. What we got is saved here:
/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PreER9/H1/Measurements/DARMOLGTFs/2016-06-29_H1_PCAL2DARMTF_4to1200Hz.xml
As a first guess it looks like the unstable mode seen at 18039Hz is a supernyquist ETMY mode. This is from the frequency shift of this mode over yesterday evening. In the first figure the change in frequency of a bunch of peaks around 18kHz is compared to the known set of 15200Hz drumhead mode frequency shift. The frequency shift is -2.5 times the ETMY mode at sample point n=50 (we'd expect -2.6 times) and then they both aproach zero frequency shift towards n=67 (ETMY ring heater was stepped). This may be contrasted to the 18002Hz mode that appears to be a subnyquist ETMY mode. Assuming this preliminary analysis is correct this mode is a 47497Hz ETMY mode.
The new HF DCPD channels allow us to see markedly more of the ringup curve (about one order of magnitude). See the comparison between the original and HF DCPD channels in the second figure.
Several modes are going through an amplitude transient. Tracking of some of the largest modes in the 18kHz area are shown in the last 2 figures showing mode amplitude vs time. Strangely the second peak at 18056 also looks like an supernyquist 47480Hz ETMY mode. The difference in amplitude transient makes me think it is real.
The arm mode spacing simulation from TCS indicates this instability occured when the mode spacings were X - 4950Hz, Y - 5015Hz. The chnage of ETMY ring heater dropped the mode spacing of ETMY by 5-10Hz. There was not instability in the last lock at this mode spacing (see the last image )however it was short.