The first attached screenshot is a series of DCPD spectra from the last several months. When we first turned on the HPO, the noise in DARM did not get any worse. The noise stayed similar to O1 until July 16th, the lump from 200-1kHz started to appear in late July, when we started a series of changes in alignment and TCS to keep us stable with a decent recycling gain at 50 Watts.
PMC
Last night, Gabriele pointed out that the noise in DARM is coherent with the PMC HV. The HV readback had been mostly some white noise (probably ADC noise), until the last few weeks, but has been getting noisier so that some of the things jitter peaks show up in it now (SEcond attached screenshot, colors corespond to the dates in the DCPD spectrum legend. This may be related to the problem described in LLO alogs 16186 and 15986. The PMC transmision has been degrading since July, which could be a symptom of misalingment. Since July, the REFL power has nearly doubled from 22 to 38 Watts, while the transmission has dropped 28%. The PMC sum has also dropped by 4%, roughly consistent with the 3% drop in the power out of the laser. Peter and Jason are planning on realigning the PMC in the morning, so it will be interesting to see if we see any difference in the HV readback.
TCS + PRC alignment:
The other two main changes we have had in this time are changes in the alignment through the PRC, and changes to TCS. These things were done to improve our recycling gain and stability without watching the noise impact carefully. In July we were using no offsets in the POP A QPDs. We started changing the offsets on August 1st, after the lumpy noise first appeared around July 22nd. We have continued to change them every few weeks since then, but generally moving in the same direction.
The only TCS change that directly coresponds to the dates when our noise got worse was the reduction of ETMY RH from 0.75 W each on July 22nd, the other main TCS changes happened September 10th. It would be nice to undo some of these changes before turning off the HPO, even if it means reduing the power to be stable.
The HVMon signal (PMC length) shows a peak at about 600Hz/rtHz. We don't this is an indication of frequency noise from the laser, but rather an error point offset picked up in the PMC PDH signal. As such this is length noise added to the PMC and suppressed by the cavity storage time. Assuming this factor is about 1/10000, we would still get ~100mHz/rtHz which are modulated onto the laser frequency. Seems a lot.
The HVMon measure 1/50 of the voltage send to the PZT. With no whitening this is not very sensitive.
After the PSL team adjusted the PMC alignment the ~400Hz peaks are no longer visible in the HVMon spectrum. The coherence is gone as well—except for the 1kHz peak.
About the PMC:
1st screenshot shows the small improvement in DARM we got after the PMC realignment. While the coherence with PMC HV may be gone, it might just be that the PMC HV signal is now burried in the noise of the ADC. At a lockloss I went to the floor and measured HV mon, then plugged it into one of the 560s, AC coupled, 10 Hz high pass, gain of 100, and sent the output into H1:LSC-EXTRA_AI_1. We still have high coherence with this channel and DARM. (last attchment)
Also, the PMC realingment this morning did decrease the reflected power, but the transmitted power also dropped.
| refl(W) | trans(W) | sum(W) | laser power out | |
| July | 20 | 126W | 157 | 174 |
| Yesterday | 35 | 103 | 138 | 169 |
| today | 27 | 100 | 126 | 169 |
About turning the HPO on not adding noise:
Kiwamu pointed out that the uncalibrated comparison above showing that the noise did not get worse when the HPO came on was not as convincing as it should have been. This morning he and I used the pcal line hieght to scale these uncalibrated spectra to something that should be proportional to meters, although we did not worry about frequency dependent calibration. (4th screenshot) From this you can see that the noise in July was very close to what it was in March before the HPO came on, but there is some stuff in the bucket that is a little worse.
The point is made best by the last attached screenshot, which is nearly identical noise in the last good lock I could find before the HPO came on to the first decent lock after it came on. Pcal was not working at this time, so we can't use that to verify the calibration, but the input powers were similar (20Watts and 24 Watts), DCPD powers are both 20mA, and the DCPD whitening was on in both cases. (The decimation filters were changed around the same time that the HPO came on which accounts for the difference at high frequencies.)
Regarding power available to the PMC, I know this is obvious but another thing we have to consider is the ISS. Since the ISS AOM is before the PMC, it clearly also effects the amount of power available to the PMC. Peter K. can correct me if I am wrong, but it is my understanding that this happens primarily in 2 ways:
- Obviously if the ISS AOM is diffracting more light, the total amount of power availabe to the PMC decreases.
- Related to the above, the ISS AOM can cause slight distortions in the beam profile, which is dependant on how hard the AOM is driven; the harder the AOM is driven, the more it distorts the beam. These distortions cause changes to the mode matching into the PMC and therefore lower the visibility of the cavity. This has the effect of increasing the reflected power and lowering the transmitted power.
On 2016-9-21, for reasons unknown to me, the ISS control offset was changed from ~4.3 to 20. This means we are driving the ISS AOM much harder than we were previously. This then causes changes in the beam profile, which effects the PMC mode matching and lowers the cavity visibility. This is likely why, even though we have had only a 5W decrease in laser power since July, the total power into and the power transmitted by the PMC are down and the power reflected by the PMC has increased, and why we cannot return to the July PMC powers Sheila listed in her table in the above comment by simply tweaking the beam alignment into the PMC. I have attached a 120 day minute-trend of the ISS control offset (H1:PSL-ISS_CTRL_OFFSET) that shows the changes in the ISS control offset value since 2016-6-22, including the 2016-9-21 change. There are of course other reasons why the control offset changed (as can be seen on the attachment, the offset was changed several times over the last 4 months), the one on 9-21 just really stuck out.
Is there a reason why the control offset was changed so drastically? Something to do with the new ISS outer loop electronics?
