Today, I worked with Kakeru on ISS.
The problem is sort of elusive. Some time, the laser power looks fine, but after a while you may see many sharp drops in the power. Some times, the power drops happen so often that they look almost like an oscillation.
We made several measurements today and Kakeru is now putting the data together. Meanwhile, I will put my speculations on the ISS problem here.
The other day, Kakeru took the transfer function of the ISS feedback filter (he is supposed to post it soon). The filter shape itself has a large phase margin ( more than 50deg ?) at the lower UGF (~3Hz) if we assume the response of the current shunt to be flat. However, when we took the whole open loop transfer function of the ISS loop, the phase margin was only 20deg. This leads to the amplification of the intensity noise around the UGF. The attached plot is the spectrum of the ISS monitor PD. You can see a broad peak around 2.7Hz. In time series, this amplified intensity noise looks like semi-oscillation around this frequency.
Since it is very unlikely that the PD has a large phase advance at low frequencies, the additional phase advance has to be in the current shunt. We measured the response of the current shunt (see Kakeru's coming post). It had a slight high-pass shape below 100Hz (a few dB/dec). This high-pass response produces additional phase advance in the loop.
There seems to be no element to produce such a high-pass response in the current shunt circuit ( http://www.ligo.caltech.edu/docs/D/D040542-A1.pdf )
This Jamie's document shows a similar high-pass response of the current ( http://www.ligo.caltech.edu/docs/G/G030476-00.pdf page 7 )
Now the question is what causes this high-pass response. Here is my very fishy hypothesis :-)
The PA output depends not only on the pump diode current but also on the mode matching with the NPRO beam, which can be changed by the thermal lensing. If the thermal lensing is in such a condition that an increase in the temperature would reduce the mode matching, then the temperature increase associated with a pump current increase could cancel the power increase. This thermal effect would be bigger at lower frequencies. Therefore, the intensity modulation efficiency decreases at lower frequencies (high-pass behavior). If this model is true, this could explain the elusiveness of the problem, as the cancellation amount depends on the operation point of the PA.
To test this hypothesis, we can change the pump current level to see if the current shunt response changes. However, the PA current slider on the MEDM screen does not work (Rob told me it's been like this for a while). Also the front panel of the MOPA power supply does not work (Steve told me it's been like this for a while). We tried to connect to the MOPA power supply from a PC through RS-232C port, which did not work neither. We will try to fix the MEDM slider tomorrow.
I attach the transfar function of the current shunt.
There is a little gap at 10 Hz for phase, but it is a ploblem of measurement and not real one.
I attache the transfar function of ISS servo.
The 4th stage and variable gain amplifier has alomost same transfar function, so their lines pile up.
I measured the output noise of eache stage of ISS servo, and calcurated the noise ratio between input and
output of each stage.
Generaly, each noise ratio corresponds to their transfar function. This means servo filter works well, not
adding extra noise.
I attache example of them.
For 2nd stage, the noise ratio is smaller than transfar function with a few factor. This is because the
input noise is coverd by analyser's noise and ratio between output and input looks small.
This means the input noise of 2nd stage was enough small and all stage before 2nd stage work well
I fixed the broken slider to change the current of the PA.
The problem was that the EPICS database assigned a wrong channel of the DAC to the slider.
I found that the PA current adjustment signal lines are connected to the CH3 &CH4 of VMIC4116 #1. However in the database file (/cvs/cds/caltech/target/c1psl/psl.db), the slider channel (C1:PSL-126MOPA_DCAMP) was assigned to CH2. I fixed the database file and rebooted c1psl. Then the PA current started to follow the slider value.
I moved the slider back and forth by +/-0.3V while the ISS loop was on. I observed that the amount of the low frequency fluctuation of the MOPA power changed with the slider position. At some current levels, the ISS instability problem went away.
Kakeru is now taking open-loop TFs and current shunt responses at different slider settings.
MOPAs and their settings, powers of 7 years in the 40m
Yoichi and me found that the transfar function of the current shunt changed with the current of PA.
We changed PA current and fixed the unstability of ISS.
Now, laser power is stabilized finely, with band of about 1 Hz.
Yoich will post the stabilized noise spectrum.
There looks to be some non-linear relation between PA current and the TF of current shunt.
It had changed from the TF which we measured yesterday, so it might change again.
I try to write scripts to sweep PA current and measure the laser power and its rms automatically.
It will be apply for auto-adjustment of PA current.
Attached files are the transfar function of the current shunt with changing PA.
They have difference in lower frequency.
After the ISS work, I aligned the IFO and confirmed that DRMI locks with good SPOB and AS166 values.
So, near 2 of the trashcans in the control room and underneath a desk there are hundrends of ants. Is this normal?
We centerd the input of WFS QPD.
The attached plot shows that someone broke the MC_SUM_MON channel around 10:30 AM this past Wednesday the 11th. This is the EPICS monitor of the MC error point.
Come forward now with your confession and I promise that I won't let Steve hurt you.
The control room video is showing us a false ETMY image. Who worked on the ETMY camera or video today??!!
I checked a broken QPD, which was placed for PSL angle monitor, and finally I cocluded one segment of the quadrant diode was broken.
The broken segment has a offset voltage of -0.7V after 1st I-V amplifier. It means the diode segment has a current offset without any injection of light.
Tomorrow I will check a new QPD for replacement.
While continuing our efforts to lock, we noticed the procedure failed at a point it had gotten past last night: turning on the bounce/roll filters in MICH, PRC, and SRC. We checked the MICH transfer function and noticed that the unity gain point was ~10 Hz, well below the bounce modes. We tried increasing the gain but found saturation, and Rob suggested that there could be misalignment on the AP table, which Steve worked on today. We went out and found two of the PDs (ASDD133 and AS166) to be badly misaligned probably due to a bumped optic upstream. We re-aligned.
I found a strange jump of value in my data taken with tdsdata.
I couldn't find same jump in a playback of DataViewer, so I think this is a problem of tdsdata.
Be careful when you use tdsdata!
The attached file is an example of jumped data.
I try to get data with allegra and op440m, and both has same kind of jump.
(A downsampling or interpolation may be wrong.)
Rana said there is a fixed version of tdsdata in some PC, but 64bit linux may not have.
I try it tomorrow.
Quick update on my wiener filtering status:
Joe has been helping me get on the GRID, so I now have a grid certificate, and accounts on most/all of the clusters.
Joe also helped me get menkar to get S5 data so that I can do wiener filtering to the back-data.
I've been running the wiener filtering algorithm, and right now, it doesn't do anything to improve the DARM_CTRL data. I am confident that this is because something is funky in the wiener filtering algorithm somewhere. The indicator of this is that the wiener filtering calculation takes the same amount of time (~95 seconds) to calculate a filter for 64 seconds of data as for 1 hour of data (both for N = 2000 taps).
For reference, attached are my plots for the wiener filtering result for (1) 64 seconds of S5 data, and for (2) 3600 seconds of S5 data.
These plots were made using H1:DARM_CTRL as the signal to minimize, with 4 seismometers as the witness channels (EX_SEISX, EY_SEISY, LVEA_SEISX, LVEA_SEISY)
I'm working on figuring out what's going on with the filtering algorithm, and why it does work for C1:MC_L minimization, but does not work for H1:DARM_CTRL minimization.
Could not get past arm power of ~11 or so. I was suspicious of the transmon high-gain/low-gain PD handover, so I ran the matchTransMon scripts, but that did not help. I also removed the line in the cm_step script that increased the CM gain by 18dB at an arm power of 4. The gain of the CM servo will increase naturally as the power in the IFO builds up, so it may not be good to crank it right away. I tried several other CM gains, and watched the DARM loop, but still could not get past an arm power of ~10-11. I'm not sure what's wrong, but it may be that mysterious CM-servo/McWFS conspiracy, so we can try turning down the McWFS gain next time.