We've seen this before, but we need to figure out why POP22 decreases with decreased CARM offset.  If it's just a demod phase issue, we can perhaps track this by changing the demod phase as we go, but if we are actually losing control of the PRMI, that is something that we need to look into.

In other news, nice work Q!

Last night I measured our RAM offsets and looked at how those affect the PRMI situation.  It seems like the RAM is not creating significant offsets that we need to worry about.

Words here about data gathering, calibration and calculations.

Step 1:  Lock PRMI on sideband, drive PRM at 675.13Hz with 100 counts (675Hz notches on in both MICH and PRCL).  Find peak heights for I-phases in DTT to get calibration number.

Step 2:  Same lock, drive ITMs differentially at 675.13Hz with 2,000 counts.  find peak heights for Q-phases in DTT to get calibration number.

Step 3:  Look up actuator calibrations.  PRM = 19.6e-9/f^2 meters/count and ITMs = 4.68e-9/f^2 meters/count.  So, I was driving PRM about 4pm, and the ITMs about 20pm.

Step 4:  Unlock PRMI, allow flashes, collect time series data of REFL RF siganls.

Step 5: Significantly misalign ITMs, collect RAM offset time series data.

Step 6: Close PSL shutter, collect dark offset time series data.

Step 7: Apply calibration to each PD time series.  For each I-phase of PDs, calibration is (PRM actuator / peak height from step 1).  For each Q-phase of PDs, calibration is (ITM actuator / peak height from step 2).

Step 8:  Look at DC difference between RAM offset and dark offset of each PD.  This is the first 4 rows of data in the summary table below.

Step 9:  Look at what peak-to-peak values of signals mean.  For PRCL, I used the largest pk-pk values in the plots below.  For MICH I used a calculation of what a half of a fringe is - bright to dark.  (Whole fringe distance) = (lambda/2), so I estimate that a half fringe is (lambda/4), which is 266nm for IR.  This is the next 4 rows of data in the table.

Step 10: Divide.  This ratio (RAM offset / pk-pk value) is my estimate of how important the RAM offset is to each length degree of freedom. 

Summary table:

Plots (Left side is several PRMI flashes, right side is a zoom to see the RAM offset more clearly):

I realized today that I had been plotting the wrong thing for all of my transfer functions for the last few weeks! 

The "CARM offsets" were correct, in that I was moving both ETMs, so all of the calculations were correct (which is good, since those took forever). But, in the plots I was just plotting the transfer function between driving ETMX and the given photodiode.  But, since just driving a single ETM is an admixture of CARM and DARM, the plots don't make any sense.  Ooops.

In these revised plots (and the .mat file attached to this elog), for each PD I extract from sigAC the transfer function between driving ETMX and the photodiode.  I also extract the TF between driving ETMY and the PD.  I then  sum those two transfer functions and divide by 2.  I multiply by the simple pendulum, which is my actuator transfer function to get to W/N, and plot.

The antispring plots don't change in shape, but the spring side plots do.  I think that this means that Rana's plots from last week are still true, so we can use the antispring side of TRX to get down to about 100 pm.

Here are the revised plots:

I've taken a first stab at this. Through various means, I've made an estimation of the total noise RMS of each error signal, and plotted a shaded region that shows the range of values the error signal is likely to take, when the IFO is statically sitting at one CARM offset. 

I have not included any effects that would change the RMS of these signals in a CARM-offset dependent way. Since this is just a rough first pass, I didn't want to get carried away just yet. 

For the transmission PDs, I measured the RMS on single arm lock. I also measured the incident power on the QPDs and thorlabs PDs for an estimate of shot noise, but this was ridiculously smaller than the in-loop RIN. I had originally though of just plotting sensing noise for the traces (i.e. dark+shot), because the amount of seismic and frequency noise in the in-loop signal obviously depends on the loop, but this gives a very misleading, tiny value. In reality we have RIN from the PRC due to seismic noise, angular motion of the optics, etc., which I have not quantified at this time. 

So: for this first, rough, pass, I am simply multiplying the single transmission noise RMSs by a factor of 10 for the coupled RMS. If nothing else, this makes the SqrtInv signal look plausible when we actually practically find it to be plausible. 

For the REFL PDs, I misaligned the ITMs for a prompt PRM reflection for a worst-case shot noise situation, and took the RMS of the spectra. (Also wrote down the dark RMSs, which are about a factor of 2 lower). I then also multiplied these by ten, to be consistent with the transmission PDs. In reality, the shot noise component will go down as we approach zero CARM offset, but if other effects dominate, that won't matter. 

Enough blathering, here's the plot:

Now, in addition to the region of linearity/validity of the different signals, we can hopefully see the amount of error relative to the desired CARM offset. (Or, at least, how that error qualitatively changes over the range of offsets)

This suggests that we MAY be able to hop over to a normalized RF signal; but this is a pretty big maybe. This signal has the response of the quotient of two nontrivial optical plants, which I have not yet given much thought to; it is probably the right time to do so...

[EricQ, Jenne]

The first half of our evening was spent working on CARM and DARM in PRFPMI, and then we moved on to the PRMI part.

I moved the DARM ALSdiff -> TransDiff transition to be after the CARM ALScomm -> SqrtInvTrans transition in the carm_cm_up script.  After I did that, I succeeded every time (at least  10?  We did it many times) to get both CARM and DARM off of the ALS signals. 

We tried for a little while looking at transitioning to REFL11 normalized by the sum of the transmissions, but we kept losing lock.  We also several times lost lock at arm powers of a few, when we thought we weren't touching the IFO for any transitions.  Looking at the lockloss time series did not show any obvious oscillations in any of the _IN1 or _OUT channels for the length degrees of freedom, so we don't know why we lost lock, but it doesn't seem to be loop oscillations caused by changing optical gain.  Also, one time, I tried engaging Rana's "Lead 350" filter in FM7 of the CARM filter bank when we were on sqrtInvTrans for CARM, and the arm powers were around a few, but that caused the transmission signals to start to oscillate, and after one or two seconds we lost lock.  We haven't tried the phase lead filter again, nor have we tried the Boost2 that is in FM8. 

We increased the REFL11 analog gain from 0dB to 12dB, and then reset the dark offsets, but still weren't able to move CARM to normalized REFL11. Also, I changed the POP22 demod phase from 159 degrees to 139 degrees. This seems to be where the signal is maximized in the I-phase, while the arms are held off resonance, and also partway up the resonance peak. 

We then decided that we should go back to the PRMI situation before trying to reduce the CARM offset further.  We can robustly and quickly lock the PRMI on REFL33 while the arms are held off resonance with ALS.  So, we have been trying to acquire on  REFL33 I&Q, and then look at switching to REFL 165 I&Q.  It seems pretty easy to get PRCL over to REFL165 I (while leaving MICH on REFL33 I).  For REFL33, both matrix elements are +1.  For PRCL on REFL165, the matrix element is -0.08.  We have not successfully gotten MICH over to REFL 165 ever this evening. 

We went back and set the REFL165 I&Q offsets so that the outputs after the demod phase were both fluctuating around 0.  I don't know if they were around +/-100 because our dark offsets were bad or what, but we thought this would help.  We were still able to get PRCL transitioned no problem, but even after remeasuring the MICH REFL33 vs. REFL165 relative gains, we still can't transition MICH.  It seems like it's failing when the REFL33Q matrix element finally gets zeroed out, so we're not really getting enough signal in REFL165Q, or something like that, and throughout the rest of the transition we were depending entirely on REFL33Q. 

So. Plan:

• Get PRMI on REFL165 while arms are held off resonance. 
  • May require PRCL-MICH FF decoupling, by combining error signals?
  • May require looking back at simulations to see what we expect the relative gains and signs to be.
• Look at CARM loop stability in simulation for REFLDC, REFL11, and normalized REFL11.  Is there a stable loop path from about 100pm down to 0pm on normalized REFL11?

 This is looking very useful. It will be useful if you can upload some python code somewhere so that I can muck with it.

I would guess that the right way to determine the trans RMS is just to use the single arm lock RIN and then apply that as RIN (not pure TR RMS) to the TR signals before doing the sqrt operation.

I changed the carm_cm_up.sh script so that it requires fewer human interventions.  Rather than stopping and asking for things like "Press enter to confirm PRMI is locked", it checks for itself.  The sequence that we have in the up script works very reliably, so we don't need to babysit the first several steps anymore. 

Another innovation tonight that Q helped put in was servoing the CARM offset to get a certain arm power.  A failing of the script had been that depending on what the arm power was during transition over to sqrtInvTrans, the arm power was always different even if the digital offset value was the same.  So, now the script will servo (slowly!!) the offset such that the arm power goes to a preset value.

The biggest real IFO progress tonight was that I was able to actually measure the CARM and DARM loops (thanks ChrisW!), and so I discovered that even though we are using (TRX-TRY)/(TRX+TRY) for our IR DARM error signal, we needed to increase the digital gain for DARM as the CARM offset was reduced.  For ALS lock and DC trans diff up to arm powers of 3, we use the same ol' gain of 6.  However, between 3 - 6, we need a gain of 7.  Then, when we go to arm powers above 6 we need a gain of 7.5.  I was also measuring the CARM loop at each of these arm powers (4, 6, 7, 8, 9), but the gain of 4 that we use for sqrtInvTrans was still fine. 

So, the carm_cm_up script will do everything that it used to without any help (unless it fails to find IR resonance for ALS, or can't lock the PRMI, in which case it will ask for help), and then once it gets to these servo lines to slowly increase the arm power and increase the DARM gain, it will ask you to confirm before each step is taken.  The script should get you all the way to arm powers of 9, which is pretty much exactly 100pm according to Q's Mist plot that is posted.

The CARM and DARM loops (around the UGFs) don't seem to be appreciably changing shape as I increase the arm powers up to 9 (as long as I increase the DARM loop gain appropriately).  So, we may be able to go a little bit farther, but since we're at about 100pm, it might be time to look at whether we think REFL11 or REFLDC is going to be more promising in terms of loop stability for the rest of the way to resonance.

Here are some plots from this evening. 

First, the time I was able to get to and hold at arm powers of 9.  I have a striptool to show the long time trends, and then zooms of the lockloss.  I do not see any particular oscillations or anything that strikes me as the cause for the lockloss.  If anyone sees something, that would be helpful.

This next lockloss was interesting because the DARM started oscillating as soon as the normalization matrix elements were turned on for DARM on DC transmissions.  The script should be measuring values and putting in matrix elements that don't change the gain when they are turned on, but perhaps something didn't work as expected.  Anyhow, the arm powers were only 1ish at the time of lockloss.  There was some kind of glitch in the DARM_OUT (see 2nd plot below, and zoom in 3rd plot), but it doesn't seem to have caused the lockloss.

[Jenne, Diego]

We spent the afternoon working on the new scan for IR resonance script.  It is getting much closer, although we need to work on a plan for the fine scanning at the end - so far, the result from the wavelet thing mis-estimates the true peak phase, and so if we jump to where it recommends, we are only at about half of the arm resonance.  So, in progress, but moving forward.

Tonight we repeated the process of reducing the CARM offset and measuring the DARM loop gain as we went.  I'm not sure if I just had the wrong numbers yesterday, or if the gains are changing day-by-day.  The gains that it wanted at given arm buildups were constant throughout this evening, but they are about a factor of 2 higher than yesterday.  If they really do change, we may need to implement a UGF servo for DARM.  New gains are in the carm_cm_up script.

We also actually saved our DARM loop measurements as a function of CARM offset (as indicated by arm buildups).  The loop stays the same through arm powers of 4.  However, once we get to arm powers of 6, the magnitude around 100 Hz starts to flatten out, and we get some weird features in the phase.  It's almost like the phase bubble has a peak growing out of it.  I saw these yesterday, and they just keep getting more pronounced as we go up to arm powers of 7, 8 and 9 (where we lost lock during the measurement).  The very last point in the power=9 trace was just before/during the lockloss, so I don't know if we trust it, or if it is real and telling us something important.  But, I think that it's time to see about getting both CARM and DARM onto a different set of error signals now that we're at about 100pm.

I spent some time trying to debug our inability to get MICH onto REFL165Q while the arms are held off with ALS, to no real success. 

I set up our usual repeatable situation of PRMI on 33 I&Q, arms held off with ALS. I figured that it may help to first sideband lock on REFL55, since 165 is looking for the f2 sidebands and maybe there is some odd offset between the locking points for f1 and f2 or other weirdness. 

REFL 55 settings:

Demod angle 98->126 (was previously set for PRY locking)

PRCL = 0.5 * REFL55 I (UGF of ~200 Hz) (FM gain unchanged from REFL33 situation of -0.02)

MICH = 0.125 * REFL55 Q (UGF of ~60Hz) (same FM gain as 33)

Some REFL55 offset adjusting had to be done in order to not disturb the 33-initiated lock when handing off. 

I also adjusted POP110 phase to zero the Q when locked, and switched the triggering over to 110I

The PRMI can acquire lock like this with arms held off with ALS, no problem. 

Here, I tried to hop over to 165. PRCL was no problem, needing a +1 on 165I. However, I had no success in handing off MICH.   I twiddled many knobs, but none that provably helped. 

I saw indications that the sensing angle in 165 is small (~20deg), which is not consistent with current knowledge of the cavity lengths. We last interferometrically measured the PRC length by letting the PRMI swing and looking at sideband splitting in POP110. At LLO, they did a length measurement by looking at demod angle differences in PRMI carrier vs. sideband locking. (alog8562) This might be worth checking out...

Short report: Further frustrated by 165 tonight. The weird thing is, the procedure I'm trying with the arms held off on ALS (i.e. excitation line in MICH and PRCL, adjust relative gains to make the signs and magnitudes mach, ezcastep over) works flawlessly with the ETMs misaligned. One can even acquire SB PRMI lock on 165 I&Q, with 80-90 degrees of demod angle between MICH and PRCL. The only real difference in REFL55 settings for misaligned vs. ALS-offset arms is an extra factor of two in the FM gains to maintain the same UGF, so I hoped that the matrix elements for 165 with misaligned arms would hold for ALS-offset arms. 

Alas, no such fortune. I still have no clear explanation for why we can't get MICH on 165Q with the arms held off on ALS. 

I also gave a quick try to measuring the PRCL->REFL55 demod phase difference between carrier and sideband lock (with arms misaligned), and got something on the order of 55 degrees, which really just makes me think I wasn't well set up / aligned, rather than actually conveying information about the PRC length...

[Koji, Jenne, Diego]

Summary:  We really don't have any MICH signal in REFL 165.  Why is still a mystery.

We made several transfer function measurements while PRMI was locked on REFL33 with the arms held off resonance, and compared those to the case where the ETMs are misaligned.  We fine-tuned the REFL165 demod phase looking at the transfer function between 10-300 Hz (using bandpassed white noise injected in the MICH FF filter bank and looking at REFL165Q), rather than just a single line.  We did that at CARM offset of 3 counts (ALS locked), and then saw that as we reduced the CARM offset, the coherence between MICH injection and REFL165Q just goes down.  Any signal that is there seems to be dominated by PRCL. 

So, we're not sure why having the arms eats the MICH 165 signal, but it does.  Everyone should dream tonight about how this could happen. 

Koji suggested that if the signal is just lost in the noise, perhaps we could increase our modulation depth for 55MHz (currently at 0.26, a pretty beefy number already).  Alternatively, if instead the problem is that the MICH signal has rotated to be in line with the PRCL signal, there may be no hope (also, why would this happen?).

Anyhow, we'd like to understand why we don't have any MICH signal in REFL165 when the arm cavities are involved, but until we come up with a solution we'll stick with REFL33 and see how far that gets us. 

The only really worthwhile plot that I've got saved is the difference in these transfer functions when PRMI-only locked and PRMI+arms locked.  Green is PRMI-only, with the demod phase optimized by actuating on PRM and minimizing the peak in the Q signal.  Blue is PRMI with the arms held off resonance using the ALS signals, with the demod phase set again, in the same way.  We were expecting (at least, hoping) that the blue transfer function would have the same shape as the green, but clearly it doesn't.  The dip that is around 45 Hz can be moved by rotating the demod phase, which changes how much PRCL couples into the Q phase.  Weird.  At ~3nm we had somewhat reasonable coherence to RELF165Q, and were able to pick -102deg as the demod phase where the dip just disappears.  However, as the CARM offset is reduced, we lost coherence in the transfer functions.

[Diego, Jenne]

The script is moving forward and we feel we are close, however we still have a couple of issues, which are:

1) some python misbehaviour between the system environment and the anaconda one; currently we call bash commands within the python script in order to avoid using the ezca library, which is the one complaining;

2) the fine scan is somewhat not so robust yet, need to investigate more; the main suspects are the wavelet parameters given to the algorithm, and the Offset and Ramp parameters used to perform the scan.

Here is an example of a best case scenario, with 20s ramp and 500 points:

Earlier today, I did some simulations that suggested that PRC lengths on the order of a cm from our current estimated one could result in degenerate PRCL and MICH signals in REFL165 at 3nm CARM offset. I attempted more demod-angle derived cavity PRC length measurements with REFL11 and REFL55, but they weren't consistent with each other...

In any case, adding dual recycling, even with a SRC length off by 1cm in either direction, doesn't seem to exhibit the same possibility, so I spent some time tonight seeing if I could make any progress towards DRMI locking. 

I was able to lock SRY using AS55 in a very similar manner to PRY, after adjusting the AS55 demod angle to get the error signal entirely in I. I used this configuration to align the SRM to the previously aligned BS and ITMY. Oddly, I was not able to do anything with SRX as I had hoped; the error signal looks very strange, looking more like abs(error signal). 

I then was able to lock the SRMI on AS55 I & Q, the settings have been saved in the IFO configure screen.  I've used AS55Q for PRMI locking with a gain of -0.2, so I started with that; the final gain ended up being -0.6. PRMI/PRY gain for prcl is something like 0.01, so since I used a gain of 2 for locking SRX, I started the SRCL gain around 0.02, the final gain ended up being -0.03. I basically just guessed a sign for AS110 triggering. Once I lucked upon a rough lock, I excited the PRM to tune the AS55 angle a few degrees; it was luckily quite close already from the SRY adjustment. AS110 needed a bigger adjustment to get the power into I. (AS55: -40.25->-82.25, AS110: 145->58, but I put AS55 back for PRMI)

I briefly tried locking the DRMI, but I was really just shooting in the dark. I went back and measured various sensing amplitudes/angles in SRMI and PRMI configurations; I'm hoping that I may be able to simulate the right gains/angles for eventual DRMI locking.

I'm leaving the iFO now. It is left with the IR arm mode.

I pretty much messed up LSC configurations for my DRMI locking. If one needs to recover the previous setting, use burtrestore.
I have all records of my LSC settings, so you don't need to preserve it. (Of course we can always use the hourly snapshots
to come back this DRMI setting)

Continued from ELOG 10659

DRMI locking

Following Jenne's elog entry in Aug 2013 (9049), DRMI was configured and locked. The lock was stable, indefinite, and repeatitive.

- DRMI Configuration

Demod phases has not been changed from PRMI

REFL11: WTN 0dB PHASE 21deg, REFL11I x0.1 -> PRCL
REFL55: WTN 21dB PHASE 25deg, REFL55Q x1 -> MICH, REFL55I x1 -> SRCL

AS110 phase was adjusted to maximize Q during the lock: +1deg (AS110Q_ERR was +4400 ~ +5500)

PRCL: GAIN -0.05 FM4/5 ON, Triggered FM 2/3/6/9, Servo trigger: POP22I 20up 10down, No Normaization.
MICH: GAIN +1 FM4/5 ON, Triggered FM 2/3/6/9, Servo trigger: POP22I 20up 10down, No Normaization.

SRCL: GAIN +2 FM4/5 ON, Triggered FM2/3/6/8/9, Servo trigger: AS110Q up 500 down 5, No Normaization.
(FM8 was set to be x2.5 flat gain such that the gain is increased after the lock)

MICH actuation is still BS+PRM and does not include SRCL decoupling yet.
This should be fixed ASAP.

DRMI Calibration

Let's use these entries 

SRM: http://nodus.ligo.caltech.edu:8080/40m/10664
SRM = (19.0 +/- 0.7) x 10 ^-9/ f²

PRM: http://nodus.ligo.caltech.edu:8080/40m/8255
PRM:  (19.6 +/- 0.3) x 10 ^-9 / f² m/counts

BS/ITMs http://nodus.ligo.caltech.edu:8080/40m/8242
BS     = (20.7 +/- 0.1)    x 10 ^-9 / f² m/counts
ITMX = (4.70 +/- 0.02)  x 10 ^-9/ f² m/counts
ITMY = (4.66 +/- 0.02) x 10 ^-9/ f² m/counts

- PRCL Calibration

Lock-in oscillator module 675.13Hz 100 -> +1 PRM

Measurement bandwidth 0.1Hz -> Signal power BW 0.471232 (FLATTOP window)

C1:SUS-PRM_LSC_IN1: 97.45 cnt/rtHz => 4.19 pm/rtHz

REFL11I: 12.55   cnt/rtHz => 3.00e12 cnt/m
REFL11Q:  0.197  cnt/rtHz => 4.70e10 cnt/m => 0.90 deg rotated! (GOOD)

REFL33I:  1.63   cnt/rtHz => 3.89e11 cnt/m
REFL33Q:  0.196  cnt/rtHz => 4.68e10 cnt/m => 8.32 deg rotated!

REFL55I:  0.0495 cnt/rtHz => 1.18e10 cnt/m
REFL55Q:  0.548  cnt/rtHz => 1.31e11 cnt/m => 84.8 deg rotated! (WHAT!)

REFL165I: 1.20   cnt/rtHz => 2.86e11 cnt/m
REFL165Q: 0.458  cnt/rtHz => 1.09e11 cnt/m => 20.9 deg rotated!

- MICH Calibration

Lock-in oscillator module 675.13Hz 100 -> -1 ITMX +1 ITMY

Measurement bandwidth 0.1Hz -> Signal power BW 0.471232 (FLATTOP window)

C1:SUS-ITMX_LSC_IN1: 121.79 cnt/rtHz => 1.26pm/rtHz
C1:SUS-ITMY_LSC_IN1: 121.79 cnt/rtHz => 1.25pm/rtHz

AS55Q:   12.45   cnt/rtHz => 4.96e12 cnt/m (STRONG)

REFL11I:  0.0703 cnt/rtHz => 2.80e10 cnt/m
REFL11Q:  0.0142 cnt/rtHz => 5.66e09 cnt/m => 78.5 deg rotated! (WHAT!)

REFL33I:  0.0473 cnt/rtHz => 1.88e10 cnt/m
REFL33Q:  0.0291 cnt/rtHz => 1.16e10 cnt/m => 58.4 deg rotated!


REFL55I:  0.00668cnt/rtHz => 2.66e09 cnt/m
REFL55Q:  0.0261 cnt/rtHz => 1.04e10 cnt/m => 14.4 deg rotated! (OK)

REFL165I: 0.0233 cnt/rtHz => 9.28e09 cnt/m
REFL165Q: 0.0512 cnt/rtHz => 2.04e10 cnt/m => 24.5 deg rotated! (GOOD)

- SRCL Calibration

Lock-in oscillator module 675.13Hz 100 -> SRM

Measurement bandwidth 0.1Hz -> Signal power BW 0.471232 (FLATTOP window)

C1:SUS-SRM_LSC_IN1: 121.77 cnt/rtHz => 5.08pm/rtHz

AS55I:    0.256   cnt/rtHz => 5.05e10 cnt/m
AS55Q:    0.3498  cnt/rtHz => 6.90e10 cnt/m

REFL11I:  0.00624 cnt/rtHz => 1.23e09 cnt/m
REFL11Q:  0.00204 cnt/rtHz => 4.02e08 cnt/m

REFL33I:  0.00835 cnt/rtHz => 1.65e09 cnt/m
REFL33Q:  0.0659  cnt/rtHz => 1.30e10 cnt/m


REFL55I:  0.0201  cnt/rtHz => 3.97e09 cnt/m
REFL55Q:  0.01505 cnt/rtHz => 2.97e09 cnt/m

REFL165I: 0.0238  cnt/rtHz => 4.69e09 cnt/m
REFL165Q: 0.0247  cnt/rtHz => 4.87e09 cnt/m

DRMI Openloop measurements
Servo filter TF measurements

The UGFs were ~250Hz for PRCL and ~100Hz for MICH, and ~250Hz for SRCL, respectively.
MICH showed (presumably) crosscoupling related peak ~350Hz. SRCL had small deviation from the model.
This may also be related to the cross couplig.

The OLTF was modelled by the servo and violin filters TF from foton, estimated TF of the AA/AI filters, and the constant time delay.

Displacement spectra measurement

- PRCL

The OLTF compensation was not actually succesfull at 300Hz, but otherwise the situation is very similar to the one with PRMI.

- MICH

Again the servo compensation at 300Hz was not successful. If we believe that AS55Q is the best MICH sensor, the out-of-loop
noise level of MICH was quite similar to the one in PRMI. We should try to use AS55Q for DRMI MICH for investigation purpose
to see which REFL signal has the best MICH quality. REFL165 seems to be iproved in the signal amplitude. Can we use this
for locking now?

- SRCL

It is in fact difficult to tell what is the correct out-of-loop noise level. AS55I has too much contamination from MICH and is not indicating
useful info. This measurement should be tried once the sensor diagonalization is done.

REFL55I is not seeing anything real abobe 30Hz. We should be able to reduce the UGF and the servo gain.

The absolute motion level of SRCL is something similar to PRCL, rather than MICH.

SRM Calibration

After the DRMI measurements on Friday, SRY cavity was locked in order to compare ITMY and SRM actuators.

SRY cavity was locked with AS55Q ->  SRM servo with gain of +10?
(My memory is fading. I tried +50 and noticed it was saturated at the limiter. So I thought it was 10)

Then the transfer functions between SRM->AS55Q TF and ITMY->AS55Q TF were measured.

The ratio between two transfer functions was obtained as seen in the second attachment.
The average at f<100Hz was 4.07 +/- 0.15. Therefore the calibration is ... as you can find below

SRM: http://nodus.ligo.caltech.edu:8080/40m/10664
SRM = (19.0 +/- 0.7) x 10 ^-9/ f²

PRM: http://nodus.ligo.caltech.edu:8080/40m/8255
PRM:  (19.6 +/- 0.3) x 10 ^-9 / f² m/counts

BS/ITMs http://nodus.ligo.caltech.edu:8080/40m/8242
BS     = (20.7 +/- 0.1)    x 10 ^-9 / f² m/counts
ITMX = (4.70 +/- 0.02)  x 10 ^-9/ f² m/counts
ITMY = (4.66 +/- 0.02) x 10 ^-9/ f² m/counts

Given the checkout of the aLIGO BBPDs happening (aLOG link), wherein the PDs were acting funny, and Koji has made some measurements determining that intermodulation/nonlinearity of circuitry can corrupt 3F signals, I've made a similar measurement of the RF spectra of REFL165 when we're locked on DRMI using 1F signals. Maybe this could give us insight to our bad luck using REFL165...

In essence, I plugged the RF output of the PD into an AG4395, through a 10dB attenuator and downloaded the spectrum. I also did REFL33 as a possible comparison and because why not. The attached plots have the 10dB accounted for; the text files do not. 

REFL165 (Exposed PCB BBPD):

(What is all that crap between 8 and 9 fmod?)

REFL33 (Gold Box resonant RFPD):

If you look at the intermodulation at 14 (4+10) and 16 (6+10), 15 (5+10) would make any problem, thanks to the notch at 1f and 5f.

BUT, this absolute level of 165MHz is too tiny for the demodulator. From the level of the demodulated signal, I can say REFL165 has
too little SNR. We want to amplify it before the demodulator.

Can you measure this again with a directional coupler instead of the direct measurement with an attenuator?
The downstream has bunch of non-50Ohm components and may cause unknown effect on the tiny 165MHz signal.
We want to measure the spectrum as close situation as possible to the nominal configuration.

90MHz crap is the amplifier noise due to bad power bypassing or bad circuit shielding.

I have no comment on REFL33 as it has completely different amplification stages.

Green beatnotes recovered.

It was just a matter of aligning the arm greens and PSL greens on the PSL table. I suppose something knocked the PSL alignment out of whack... I was also able to simultaneously see the green beatnote and IR beatnote respond to Yend laser temperature. 

Locked arms on POX/POY, checked RMS of ALS-BEAT[X/Y]_FINE_PHASE_OUT_HZ channels. 

• ALSY: 300Hz RMS
• ALSX: 700Hz RMS

These seem fine. Locked CARM and DARM on ALS, found IR resonances. 

ALS is back in business 

Now that I have followed the chain, the PD signal is indeed being amplified at the LSC rack. It goes into a ZFL-1000LN+ amplifier (~23dB gain at 165MHz and 15V supply), followed by a SHP-100 high pass filter, and then enters the RF IN of the demod board. 

I repeated the measurement in two spots.

First, I took a spectrum of the RF MON of the REFL165 demod board during DRMI lock; this was input-referred by adding 20dBm. 

Second, I inserted a ZFDC-10-5 coupler between the high pass and the RF input of the demod board. This was input-referred by adding 10dBm. 

My calibration isn't perfect; the peaks above the high pass corner seem to be different by a consistent amount, but within a few dBm. 

Thus, it looks like the demod board is getting a little under -40dBm of 165MHz signal at its input. 

 Tonight I tried some more tests on the script; it seems to work better, with both performance and robustness improved, although the Xarm behaved badly almost all the time. I did not perform all the tests I wanted because the ALS lock was pretty unstable tonight (not only because of the X arm), with more than a few lock losses; after the last lock loss, however, I couldn't restore the Xarm. I'll do some more tests as soon I can recover it, or post the result of the first batch of tests.

In addition, I encountered the following error multiple times, but I have no idea about what could it be:

Thu Nov 06 02:00:13 PST 2014
medmCAExceptionHandlerCb: Channel Access Exception:
Channel Name: Unavailable
Native Type: Unavailable
Native Count: 0
Access: Unavailable
IOC: Unavailable
Message: Virtual circuit disconnect
Context: fb.martian.113.168.192.in-addr.arpa:5064
Requested Type: TYPENOTCONN
Requested Count: 0
Source File: ../cac.cpp
Line number: 1214
 

Where is the PD out spectrum measured with the coupler???

EDIT on X arm: I found different settings in C1SUS_ITMX, with respect to ETMX, ITMY and ETMY (namely LSC/DAMP is OFF and LSC/BIAS is ON); I don't know if this is intended or for some reason ITMX was not recovered properly after the lock loss, so I didn't change anything, but it may be worth looking into that.

Still no luck in recovering the X arm, I am giving up for tonight; honestly I didn't try many things, as I don't know well the system and didn't want to mess things up.

Preliminary results so far:

I confirm that the best settings for the ramp of the ALS scan are 20s and 500 points; this causes however the script to be fairly slow (80s for the scan/data collection, 7s for the coarse peak finding, 17s for the fine peak finding, total ~2 min); in the best cases the TR*_OUT obtained is around 0.90, as shown in the first plot (early in the evening, all the following plots are in chronological order, if that can help finding the reason for the X arm misbehaviour...):

However, after a few minutes somehow the TR*_OUT went down a bit, without any kind of intervention; also, it is visible the instability of the X arm:

Even when X arm was somewhat stable, its performance and robustness were (far) worse than the Y arm ones:

The following plot shows (about the Y arm only) that there is still some margin, as the maximum value of TRY_OUT is not completely kept at the end of the procedure:

Finally the last plot I managed to obtain, before the X arm went completely crazy...

The next step, after obviously figuring out the X arm situation, is to try some averaging during the fine scan, I don' t know if this will improve the situation, however it shouldn't impact on the execution time. Tomorrow I'll post something more detailed on the script itself and the wavelet implementation.

 The "coupled" port of the coupler went to the AG4395 input, the output of the Highpass is connected to the "IN", and the "OUT" goes to the demod board. 

That's not what I'm asking.

Also additional cables are left connected to the signal path. I removed it.

After some enlightening conversation with Koji, we figured that the RF amplifier in the REFL165 chain is probably being saturated (the amp's 1dB compression is at +3dBm, has 23dB gain, and there are multiple lines above -20dBm coming out of the PD). I took a few more spectrum measurements to quantify the consequences, as well as a test with the highpass connected directly to the PD output, that should reduce the power into the amplifier. However, I am leaving everything hooked back up in its original state (and have removed all couplers and analyzers...)

I also took some DRMI sensing measurements. In the simple Michelson configuration, I took TFs of each ITMs motion to AS55Q to make sure the drives were well balanced. They were. Then, in the DRMI, I took swept sine TFs of PRCL, SRCL and differential ITM MICH motion to the Is and Qs of AS55 and all of the REFLs. I constrained the sweeps to 300Hz->2kHz; the loops have some amount of coupling so I wanted to stay out of their bandwidth. I also took a TF of the pure BS motion and BS-PRM MICH to the PDs. From these and future measurements, I hope to pursue better estimates of the sensing matrix elements of the DRMI DoFs, and perhaps the coefficients for compensating both SRCL and PRCL out of BS motion. 

I'm leaving analysis and interpretation for the daytime, and handing the IFO back to Diego...

The measurements I took yesterday bear this out. However, even putting the high-pass directly on the PD output doesn't reduce the signal enough to avoid saturating the amplifier.

We need to think of the right way to get the 165MHz signal at large enough, but undistorted, amplitude to the demod board. 

 The current signal chain looks like:

I previously made measurements at (3). Let's ignore that. 

Last night, I took measurements with a directional coupler at points (1) and (2), to see the signal levels before and after the amplifier. I divided the spectrum at (2) by the nominal gain of the amplifier, 23.5dB; thus if everything was linear, the spectra would be very similar. This is not the case, and it is evident why. There are multiple signals stronger than -20dBm, and the amplifier has a 1dB compression point of +3dBm, so any one of these lines at 4x, 6x and 10x fMod is enough to saturate. 

I also made a measurement at point 4 in the following arrangement, in an attempt to reduce the signal amplitude incident on the amplifier.  

 Though the signals below 100MHz are attenuated as expected, the signal at 110MHz is still too large for the amplifier. 

Minicircuits' SHP-150 only has 13dB suppression at 110MHz, which would not be enough either. SHP-175 has 31dB suppression at 110MHz and 0.82dB at 160MHz, maybe this is what we want.

I have found an SHP-150, but no SHP-175's (also, several 200's, and a couple of 500's).

Why do you say the SHP-150 isn't enough?  The blue peak at 10*fmod in your plot looks like it's at -12 dBm.  -13 dB on top of that will leave that peak at -25 dBm.  That should be enough to keep us from saturation, right?  It's not a lot of headroom, but we can give it a twirl until a 175-er comes in.  

Koji also suggests putting in a 110 MHz notch, combined with either an SHP-150 or SHP-175, although we'll have to measure the combined TF to make sure the notch doesn't spoil the high pass's response too much.

Yesterday I did some more tests with a modifies script; the main difference is that scipy's default wavelet implementation is quite rigid, and it allows only very few choices on the wavelet. The main issue is that our signal is a real, always positive symmetrical signal, while wavelets are defined as 0-integral functions, and can be both real or complex, depending on the wavelet; I found a different wavelet implementation, and I combined it with some modified code from the scipy source, in order to be able to select different wavelets. The result is the wavelet_custom.py module, which lives in the same ALS script directory and it is called by the script. In both the script and the module there the references I used while writing them. It is now possible to select almost any wavelet included in this custom module; "almost" means that the scipy code that calls the find_peaks_cwt routine is picky on the input parameters of the wavelet function, I may dig into that later. For the last tests, instead of using a Ricker wavelet (aka Mexican hat, or Derivative of Gaussian Order 2), I used a DOG(6), as it also has two lesser positive lobes, which can help in finding the resonance; the presence of negative lobes is, as I said, unavoidable. I attach an example of the wavelet forms that are possible, and in my opinion, excluding the asymmetric and/or complex ones, the DOG(6) seems the best choice, and it has provided slightly better results. There are other wavelet around, but they are not included in the module so I should implement them myself, I will first see if they seem fitting our case before starting writing them into the module. However, the problem of not finding the perfect working point (the "overshoot-like" plot in my previous elog) is not completely solved. Eric had a good idea about that: during the fine scan, the the PO*11_ERR_DQ signals should be in their linear range, so I could also use them and check their zero crossing to find the optimal working. I will be working on that.

 I think 'saturation' here is a misleading term to think about. In the RF amplifiers, there is always saturation. What we're trying to minimize is the amount of distorted waveforms appearing at 3f and 15f from the other large peaks. Usually for saturation we are worried about how much the big peak is getting distorted; not the case for us.

Here are some preliminary results from the sensing sweeps I did the other night. 

Notes:

• The analog AS55I signal chain is almost certainly busted in some way. This would also explain the odd looking error signals in SRX, and was actually hypothesized by Koji when discussing the SRX oddness. 
• I used the same mirror calibration numbers from Koji's recent Elogs to turn these into counts/m.
• MICH was excited via differential ITM motion.  I also performed a TF with BS driven MICH, with the compensating PRM output matrix in place, and it looks different, but I haven't looked too deeply into it yet. 
• The angles plotted are in regard to the analog I and Q signals (i.e., I took TFs to I_ERR and Q_ERR and then unrotated by the digital rotation angle); this is why I suspect AS55I is broken, as all of the signals are entirely in the analog Q.
• The amplitudes seem to be roughly consistent with Koji's recent observations. 
• I still need to cut out the violin-filter-corrupted data points to quote the sensing elements with error bars...

Plots!

xml files, and DttData matlab script used to generate these plots is attached. 

 Jenne and I measured the situation using a SHP-150 directly attached to the REFL165 RF output, and at first glance, the magnitude of the 165MHz signal seems to not be distorted by the amplifier. 

We will soon investigate whether 165 signal quality has indeed improved. 

ARG, I accidentally permuted the digital demod angles. This significantly weakens the argument for believing AS55I is broken... In fact, Jenne and I did some investigations this afternoon that showed that the channel is indeed working. SRX error signal strangeness remains unexplained, however. 

Also, I have yet to compensate for the gain of the violin filters; the actuator calibration numbers I used were for the SUS-LSC FMs, not the LSC FMs where I was injecting. New measurements will be taken soon, as well, since REFL165 is hopefully improved. 

Corrected plots are below. 

To further reduce the RF power at 110MHz in the REFL165 chain, I made a twin-t notch in a pomona box.

It is tuned at 110.66MHz.

The inductor is Coil Craft 5mm tunable (164-09A06SL 100-134nH).
Without the 10Ohm resister (like a usual notch), the dip was ~20dB. With this configuration, the notch of -42dB was realized.

Q >> Please measure the RF spectrum again with the notch.

I was able to lock the DRMI on 3f signals this evening, although the loops are not stable, and I can hear oscillations in the speakers.  I think the big key to making this work was the placement of the SHP-150 high pass filter at the REFL165 PD, and also the installation of Koji's 110 MHz notch filter just before the amplifier, which is before the demod board for REFL165.  These were done to prevent RF signal distortion.

DRMI 3f:   With DRMI locked on 1f (MICH gain = 1, PRCL gain = -0.05, SRCL gain = 2, MICH = 1*REFL55Q, PRCL = 0.1*REFL11I, SRCL = 1*REFL165I), I excited lines, and found the signs and values for 3f matrix elements.  I was using the same gains, but MICH = 0.5*REFL165Q, PRCL = 0.8*REFL33I and SRCL = -0.2*REFL165I.  Part of the problem is likely that I need to include off-diagonal elements in the input matrix to remove PRCL from the SRCL error signal. 

With the DRMI locked on 1f, I took a sensing matrix measurement.  I don't think we believe the W/ct of the photodiode calibration (we need to redo this), but otherwise the sensing matrix measurement should be accurate.  Since the calibration of the PDs isn't for sure, the relative magnitude for signals between PDs shouldn't be taken as gospel, but within a single PD they should be fine for comparison. 

As a side note, we weren't sure about the MICH -> ITMs balancing, so we checked during a MICH-only, and with the locking apparatus we are unable to measure a difference between 1's for both ITMs in the output matrix, and 1 for ITMX and 0.99 for ITMY.  So, we can't measure 1% misbalances in the actuator, but we think we're at least pretty close to driving pure MICH. 

We kind of expect that SRCL should only be present in the 55 and 165 PDs, although we see it strongly in all of the REFL PDs.  Also, PRCL and SRCL are not both lined up in the I-phase.  So, I invite other people to check what they think the sensing matrix looks like. 

• The excitation lines (and matching notches) were on from 12:14am (
• Nov 11 2014 08:14:00 UTC / GPS 1099728856) to 12:20am (
•  
• Nov 11 2014 08:20:00 UTC / GPS 
• 1099729216) for 360sec. 
• MICH was driven with 800 counts at 675.13 Hz, with +1*ITMY, -1*ITMX. 
• PRCL was driven with 1000 counts at 621.13 Hz with the PRM. 
• SRCL was driven with 800 counts at 585.13 Hz using the SRM. 

All 3 degrees of freedom have notches at all 3 of those frequencies in the FM10 of the filter banks (and they were all turned on).  During this time, DRMI was locked with 1f signals. 

DRMI sensing matrix:

Earlier in the evening, I also took a PRMI sensing matrix, with the PRMI locked on REFL33 I&Q.  Watch out for the different placement of the plots - I couldn't measure AS55 in the DRMI case, since cdsutils.avg freaked out if I asked for more than 14 numbers (#PDs * #dofs) at a time.

Rana, Koji and I were staring at the DRMI sensing matrix for a little while, and we aren't sure why PRCL and SRCL aren't co-aligned, and why they aren't orthogonal to MICH.  Do we think it's possible to do something to digitally realign them?  Will the solution that we choose be valid for many CARM offsets, or do we have to change things every few steps (which we don't want to do)? 

Things to work on:

* Reanalyze DRMI sensing matrix data from 12:14-12:20am. 

* Make a simulated scan of higher order mode resonances in the arm cavities.  Is it possible that on one or both sides of the CARM resonance we are getting HOM resonances of the sidebands? 

* Figure out how to make DRMI 3f loops stable.

* Try CARM offset reduction with DRMI, and / or PRMI on REFL 165.

Sensing matrix calculation using DTT + Matlab

Note: If the signal phase is, for example,  '47 deg', the phase rotation angle is -47deg in order to bring this signal to 'I' phase.

Note2: As I didn't have the DQ channels for the actuation, only the relative signs between the PDs are used to produce the radar chart.
This means that it may contain 180deg uncertainty for a particular actuator. But this does not change the independence (or degeneracy) of the signals.


=== Sensing Matrix Report ===
Test time: 2014-11-11 08:14:00
Starting GPS Time: 1099728855.0
 

== PRCL ==
Actuation frequency: 621.13 Hz
Suspension (PRM) response at the act. freq.: 5.0803e-14/f^2 m/cnt
Actuation amplitude: 20.3948 cnt/rtHz
Actuation displacement: 1.0361e-12 m/rtHz
 
C1:LSC-AS55_I_ERR_DQ 4.20e+10
C1:LSC-AS55_Q_ERR_DQ -1.91e+11
==> AS55: 1.95e+11 [m/cnt] -24.58 [deg]
C1:LSC-REFL11_I_ERR_DQ 3.17e+12
C1:LSC-REFL11_Q_ERR_DQ -8.04e+10
==> REFL11: 3.17e+12 [m/cnt] -18.20 [deg]
C1:LSC-REFL33_I_ERR_DQ 4.15e+11
C1:LSC-REFL33_Q_ERR_DQ 4.28e+10
==> REFL33: 4.17e+11 [m/cnt] -137.11 [deg]
C1:LSC-REFL55_I_ERR_DQ 1.90e+10
C1:LSC-REFL55_Q_ERR_DQ -9.91e+09
==> REFL55: 2.14e+10 [m/cnt] -58.58 [deg]
C1:LSC-REFL165_I_ERR_DQ -1.16e+11
C1:LSC-REFL165_Q_ERR_DQ -3.14e+10
==> REFL165: 1.20e+11 [m/cnt] 45.20 [deg]
 
 
== MICH ==
Actuation frequency: 675.13 Hz
Suspension (ITMX) response at the act. freq.: 1.0312e-14/f^2 m/cnt
Suspension (ITMY) response at the act. freq.: 1.0224e-14/f^2 m/cnt
Actuation amplitude: 974.2957 cnt/rtHz
Actuation displacement (ITMX+ITMY): 2.0007e-11 m/rtHz
 
C1:LSC-AS55_I_ERR_DQ 2.55e+12
C1:LSC-AS55_Q_ERR_DQ 4.51e+12
==> AS55: 5.18e+12 [m/cnt] 113.51 [deg]
C1:LSC-REFL11_I_ERR_DQ -4.84e+10
C1:LSC-REFL11_Q_ERR_DQ -4.07e+09
==> REFL11: 4.85e+10 [m/cnt] 168.06 [deg]
C1:LSC-REFL33_I_ERR_DQ 2.06e+10
C1:LSC-REFL33_Q_ERR_DQ -9.39e+09
==> REFL33: 2.26e+10 [m/cnt] -167.51 [deg]
C1:LSC-REFL55_I_ERR_DQ 2.52e+09
C1:LSC-REFL55_Q_ERR_DQ -1.02e+10
==> REFL55: 1.05e+10 [m/cnt] -107.09 [deg]
C1:LSC-REFL165_I_ERR_DQ -1.79e+10
C1:LSC-REFL165_Q_ERR_DQ -5.50e+10
==> REFL165: 5.79e+10 [m/cnt] 102.02 [deg]


== SRCL ==
Actuation frequency: 585.13 Hz
Suspension (SRM) response at the act. freq.: 5.5494e-14/f^2 m/cnt
Actuation amplitude: 1176.3066 cnt/rtHz
Actuation displacement: 6.5278e-11 m/rtHz
 
C1:LSC-AS55_I_ERR_DQ -9.90e+10
C1:LSC-AS55_Q_ERR_DQ -1.18e+11
==> AS55: 1.54e+11 [m/cnt] -76.89 [deg]
C1:LSC-REFL11_I_ERR_DQ 2.96e+08
C1:LSC-REFL11_Q_ERR_DQ 4.78e+08
==> REFL11: 5.62e+08 [m/cnt] 41.42 [deg]
C1:LSC-REFL33_I_ERR_DQ -2.93e+09
C1:LSC-REFL33_Q_ERR_DQ 1.23e+10
==> REFL33: 1.27e+10 [m/cnt] -39.63 [deg]
C1:LSC-REFL55_I_ERR_DQ 3.71e+09
C1:LSC-REFL55_Q_ERR_DQ 2.78e+09
==> REFL55: 4.63e+09 [m/cnt] 5.86 [deg]
C1:LSC-REFL165_I_ERR_DQ -1.80e+10
C1:LSC-REFL165_Q_ERR_DQ 2.68e+10
==> REFL165: 3.23e+10 [m/cnt] -26.02 [deg]
 

Demodulation phases of the day

    'C1:LSC-AS55_PHASE_R = -53'
    'C1:LSC-REFL11_PHASE_R = 16.75'
    'C1:LSC-REFL33_PHASE_R = 143'
    'C1:LSC-REFL55_PHASE_R = 31'
    'C1:LSC-REFL165_PHASE_R = 150'


The notch filter has been installed directly attached to the output of the SHP-150 at the PD output. Structurally, there is a right angle SMA elbow between the two filters; I set up a post holder under the notch pomona box to prevent torque on the PD. Via directional coupler and AG4395, we measured the output of the REFL165 RF amplifier with the PRMI locked on REFL33. 

Note, the plot below is not referred to the amplifier output, as in my previous plots; it is directly representative of the amplifier output spectrum. 

There are no RF signals being output above -28dBm, thus I am confident that we are not subject to compression distortion. 

Given the last measurements we made (ELOG 10692), I estimate that the notch has reduced the power at 110MHz by ~33dB, which is 9dB higher than the notch performance Koji measured when he made it. Maybe this could be due to the non-50Ohm impedance of the HPF distorting the tuning, or I physically detuned it when mounting it on the PD. Still, 33dB is pretty good, and may even give us room to amplify further. (ZRL-700+ instead of the ZFL-1000LN+?)

I did some simulations to see if we are susceptible to HOM resonances as we reduce the CARM offset. I restricted my search to HG modes of the Carrier+[-55,-11,0,+11,+55]MHz fields with n+m<6, and used all the real physical parameters I could get ahold of. 

In short, as I change the CARM offset, I don't see any stray resonances within 2nm of zero, either in PRFPMI or DRFPMI. 

Now, the mode matching in my simulation is not the real mode matching our real interferometer has. Thus, it can't tell us how much power we may see in a given mode, but it can tell us about our susceptibility to different modes. I.e. if we were to have some power in a certain mode coming out of the IMC, or present in the vertex, we can see what it would do in the arms. 

Since my simulation has some random amounts of power in each HOM coming into the interferometer, I simply swept the CARM offset and looked for peaks in the power of each mode. Many of the fields exhibited gentle slopes over the range, and we know we ok from 3nm->~100pm, so I made the selection rule that a "peak" must be at least 10 times as big as the minimum value over the whole range, in order to see fields that really do have CARM dependence. 

In the following plots, normalized IFO power is plotted and the locations of HOM peaks are indicated with circles; their actual heights are arbitrary, since I don't know our real mode content. However, I'm not really too concerned, since all I see is some -11MHz modes between 2-3nm of full resonance, where we have no problem controlling things... Also, all of the carrier HOMs effectively co-resonate with the 00 mode, which isn't too surprising, and I didn't include these modes in the plots.

Finally, I visually inspected the traces for all of the modes, and didn't really find anything else peeking out. 

Code, plots attached.  

Koji pointed out something to me that I think he had told me ages ago, and Rana alluded to last night:  Since I'm not tuning my "demod phase" for the sensing matrix lockins, unless I happened to get very lucky, I was throwing away most of the signal.  Lame.  

So, now the magnitude is sqrt(real^2 + imag^2), where real and imag here are the I and Q outputs of the lockin demodulator, after the 0.1Hz lowpass.  (I put in the low pass into all of the Q filter banks).  To keep the signs consistent, I did do a rough tuning of those angles, so that I can use the sign of the real part as the sign of my signal.  When I was PRMI locked, I set the phase for all things acutated by MICH to be 79deg, all things actuated by PRCL to be 20 deg, and when DRMI locked set all things SRCL to be 50deg. 

After doing this, the phases of my sensing matrix output matches Koji's careful analyses.  I don't know where the W/ct numbers I was using came from (I don't think I made them up out of the blue, but I didn't document where they're from, so I need to remeasure them).  Anyhow, for now I have 1's in the calibration screen for the W/ct calibration for all PDs, so my sensing matrices are coming out in cts/m, which is the same unit that Koji's analysis is in. (Plot for comparing to Koji is at end of entry).

While reducing the CARM offset, I left the sensing matrix lines on, and watched how they evolved.  The phases don't seem to change all that much, but the magnitudes start to decrease as I increase the arm power.

For this screenshot, the left plot is the phases of the sensing matrix elements (all the REFL signals, MICH and PRCL), and the right plot is the magnitudes of those same elements.  Also plotted is the TRX power, as a proxy for CARM offset.  The y-scale for the TRX trace is 0-15.  The y-scale for all the phases is -360 to +360.  The y-scale of the magnitude traces are each one decade, on a log scale.

Bonus plot, same situation, but the next lock held for 20 minutes at arm powers of 8.  We don't know why we lost lock (none of the loops were oscillating, that I could see in the lockloss plot).

Here are some individual sensing matrix plots, for a single lock stretch, at various times.  One thing that you can see in the striptool screenshots that I don't know yet how to deal with for the radar plots is the error bars when the phase flips around by 360 degrees.  Anyhow, the errors in the phases certainly aren't as big as the error boxes make them look.

PRMI just locked, CARM offset about 3nm, CARM and DARM on ALS comm and diff, arm powers below 1:

PRMI still on REFL33 I&Q, CARM and DARM both on DC transmissions, arm powers about 4:

CARM offset reduced further, arm powers about 6:

CARM offset reduced even more, arm powers about 7:

For this plot for comparing with Koji's analysis, I had not yet put 1's in the calibration screen, so this is still in "W"/m, where "W" is meant to indicate that I don't really know the calibration at all.  What is good to see though is that the angles agree very well with Koji's analysis, even though he was analyzing data from yesterday, and this data was taken today.  This sensing matrix is DRMI-only (no arms), 1f locking.

Bonus plot, PRMI-only sensing matrix, with PRMI held using REFL 33 I&Q:

In my previous meditations about RIN, particularly elog 10258, I was only thinking about the RIN contribution at the offset that I was currently sitting at.  Also, In elog 10258 I was comparing to the ALS signals and just said that the trans signals are better which is true, although isn't super helpful when thinking of reduced CARM offsets. 

My summary today is that I think we want to reduce the RIN in arm transmissions by a factor of 3.

Rather than dig around, I just remeasured the RIN, for both the single arm transmissions and the MC transmission.  (Data attached as .xml file)

The RMS RIN for the Xarm is 1.3e-2.  The RMS RIN for the Yarm is 8.9e-3.  The RMS RIN for MCtrans is 4.0e-3.  For the simulations below, I will use 1e-2 as an average RIN for the arms.

As an estimate of the RIN's contribution to cavity fluctuations, I divide the RIN by the slope of the CARM transmission peak.  The slope (from optickle) gives me [ delta-W / delta-m ], and the inverse of that gives me [ delta-m / delta-W ].  I multiply this by RIN, which is [ delta-W / W ] to get [delta-m / W]. 

Then, since I'm using the DC transmission signals as my error signals, I use just TRX (normalized to be 1 for single arm resonance) as my Watts.

So, in total, the traces plotted are { TRX * RIN / slope }. 

The 2 plots are the same data, one with linear-x and the other with log-x.  They both include my estimate of the cavity length fluctuations due to RIN at the arm transmission, as well as an estimate of the cavity length fluctuations if the arm RIN was as good as the MC RIN.  I also show the DRFPMI CARM linewidth (23 pm for HWHM), and 1% of that linewidth.  The last trace is 1% of the half-width of the transmission peak, at the current CARM offset.  For example, 1000 pm away from full resonance the half-width is 1000 pm and 1% of that is 10 pm. 

What we want to see here is that the solid blue line is below one of the dotted lines.  I think that using the overall linewidth (purple dotted line) isn't really the right thing to look at.  Our goal is to prevent excursions that will get too close to the resonance peak, and cause a lockloss.  A one picometer excursion is a much bigger problem (relatively) below say 100 pm, as opposed to above 100 pm.  So, I think that we should be looking at the half-width of the resonance peak at whatever the current CARM offset is (orange dotted line).  Above 25 pm, the blue line is below the orange line for all offsets plotted.  If we made the arm RIN as good as the MC RIN, that would be true down to 12-ish pm. 

We should be able to safely transition to non-normalized RF signals at 10pm or below.  This implies that (since any RF signals normalized by this RIN-y trans signal will have the RIN), we want to improve the RIN of the transmission PDs by about a factor of 3. (This will lower the blue line such that it crosses the orange dotted line at 10 pm).

So, with my last entry, I was guilty of just throwing stuff into the simulation and not thinking about physics... so I retreated to Siegman for some algebraic calculations of the additional Guoy phase accumulated by the HOMs in the arms -> their resonant frequencies -> the arm length offset where they should resonate. Really, this isn't completely precise, as I treated the arms independently, with slightly differing ETM radii of curvature, but I would expect the "CARM Arm" to behave as a sort of average of the two arm cavities in this regard. (EDIT: Also, I didn't really consider the effect of the coupled vertex cavities... so there's more to be done)

The basic idea I used was:

• Assume ITMs are effectively flat, infinite Rc
• Use 40mwiki values for ETM curvatures
• Each additional HG order adds arccos(sqrt(1 - Larm/Rc)) of Guoy phase for a one way trip down the cavity (Eqn 19.19 in Sigman)
• For each HOM order up to 5 of the carrier and first order sidebands, add the appropriate phase shift 
• fold it onto +-FSR/2 of the carrier 00 resonance, convert to m

In practice, I threw together a python script to do this all and print out a table. I've highlighted the values within 10nm, but the closet one is 3.8nm

Results:

 I modified the MC2 trans optical setup a little bit: the reflection from the QPD was not dumped and so it was hitting the wall of the black box.

I angled the QPD slightly and moved the dump so that the reflection hit it. The leakage through the 50/50 steering mirror for the QPD was already being dumped and I made sure that that one stayed dumped on its razor dump. After doing this we turned off the WFS and re-aligned the MC2 trans beam onto the QPD to zero the pit/yaw signals. 

[Jenne, Rana, Koji]

We did some thinking on what could be causing the excess RIN that we see in the arm transmissions but not in the MC transmission.  Unfortunately, I don't think we have anything conclusive yet. 

We thought about:

• Test mass oplevs
• Input tip tilt jitter
• MC motion

Oplevs

As Rana reported in elog 10708, we tuned the oplev servos for ITMX, ETMX, ITMY and ETMY.  They all now look like the new SRM oplevs that Rana described in elog 10694.  However, when we re-looked at the RIN after the oplev tuning, we did not see a noticeable change.  So, fixing up the oplevs didn't fix up the RIN.

Side notes:

• Several optics had low gains for suspos, which were increased to give Qs of ~5ish.
• When we gave ITMX a 500 count step in pitch (the same size used for all other optics in both pit and yaw), it didn't come back afterward.  This is a little disconcerting.  Rana had to move the alignment slider to get it back so that we had MICH fringing at the AS port again.

Input Tip Tilts

Koji did some work, reported in elog 10706, on how much the jitter of the input pointing tip tilts should affect us.  We don't think that they are moving enough to be the cause of the excess RIN that we see.

Mode Cleaner Motion

We see some coherence between MC2 suspit and TRX/TRY, so we suspect that the MC's motion could be causing problems. 

I looked at the WFS vs. TRX & TRY, and saw significant coherence at the 3 Hz stack resonance.  I think it's clear that the WFS can help suppress this motion more.  The WFS noise level was too bad to see any other coherence at other frequencies. (Side note:  We should consider increasing the analog WFS signal.  As Rana mentioned back in 2008 in elog 454, the signal is super small.  Increasing the analog gain could allow us to suppress motion at more frequencies, although it would be a bit of a pain.)

To try and get some more signal, I routed the IPPOS beam over to the PRM oplev temporarily.  The idea was to be able to look at the IPPOS port, but with a fast channel.  I turned off the BS/PRM HeNe, and removed the last steering mirror before the QPD.  I put in 2 Y1 steering mirrors to get the IPPOS beam across the table and pointing at IPPOS.  I took my measurements 3 times, with different beam sizes on the QPD, to try to image different gouy phases.  I used absorptive ND filters (0.6 + 0.1) to get the light level on the PD such that I had about 10,000 counts per quadrant, where 16,000 counts seemed to be the saturation point. I also reset the dark offsets of the QPD quadrants, although they were so small that I don't think it did much.  I also took out the optical lever calibration from counts to microradians, since that number isn't meaningful for what I was doing.  So, the IPPOS signals (using the PRM oplev channels) are in raw ADC counts.  The first measurement had no lens, and the beam was probably at least half the size of the QPD.  The second measurement had a lens, and the beam on the QPD was about half the original size.  The third measurement had the lens closer to the QPD, so that the beam was about 1mm on the diode.  In none of these cases do I see any significant coherence with the TRX/TRY RIN signals, except at 3 Hz. After my measurements I put the oplev back including all of the digital settings, although for now I just left the IPPOS beam dumped on a razor dump, since it wasn't being used anyway.  I need to realign IPPOS when it's not the middle of the night.

Some thoughts that we have:

• Maybe it's time to resurrect seismic feedforward on MC length, to suppress some of this 3 Hz motion?
• Maybe we should be using the MC_L path to offload some of the frequency feedback, so that we're not pushing on MC2 so hard (because if we have bad F2P coupling, this creates beam motion)

I have plots for each of my IPPOS vs. TRX/TRY measurements.  The data is attached.  For each, I looked at the transfer function between IPPOS (using the SUS-PRM_OPLEV channels) and TRX/TRY to get the 'calibration' between input beam motion and transmission RIN.  In all cases, at 3 Hz the coefficient was about 1, so in the power spectra on the right side, there is no calibration applied to the IPPOS signals. 

IPPOS vs. Transmission RIN, no lens, big beam on QPD:

(Just kidding about the other 2 plots - the elog can't handle it.  They're in the zippyzip file though, and I don't think they look qualitatively different from the no-lens case).

After Kate, Diego and I moved the seismometers to the corner for a huddle test (see elog 10711), I wanted to check the coherence between the seismometers and the arm transmissions.  

First of all, it looks like either the Guralp or the T-240 have their X and Y backwards. Diego is going to check this tomorrow. 

Between 0.9Hz - 3.5Hz, we have pretty strong coherence with the horizontal seismic channels.  Interestingly, between 8-10Hz, the Yarm has pretty strong coherence with the Z-axes of the seismometers (the coherence is only about 0.6, but it's consistent over a 2-ish Hz wide band). 

The MC transmission doesn't have as much coherence with the seismic, which surprised me.  So, we can try some FF to the MC, but we may also have to do some to the arms.

I've extended my analysis to the PRFPMI case, with the current working knowledge of radii of curvature and cavity lengths. However, losses were not included.

I do not see any HOM activity within about 20nm of the carrier TM00 resonance. 

Basically, what I did was use the standard formulae for the reflection and transmission coefficients of FB cavities viewed as compound mirrors. However, I modified the normal spatial propagation terms to include the additional Guoy phase accumulated by the HOMs. I created these coefficients for each arm individually, and then used (rX + rY)/2 as a mirror in the PRC, and used that to create the transmission coefficient for the PRFPMI as a whole, as a function of frequency offset from the carrier, spatial mode order and CARM offset. As a check, this produced the correct finesse for the carrier lock to the single arm and PRFPMI. 

Here is a PRFPMI CARM FSR of all of the fields' power transmission coefficients, up to order n+m=5. 

One can observe some split peaks. There are two causes, the biggest effect is the mismatch between ETM radii of curvatures (ETMX:59.48, ETMY:60.26):, followed by asymmetric arm length(X:37.79, Y:37.81). (I judged this by the visual change of the plot when changing different factors). 

In the following plot, I broke down the peaks by mode order:

Code, plots attached!

T-240 has a different convention than we use.  It assumes that North is aligned with the Y-axis.  Since this is the new guy, and we've been using the Guralps with North = X for many years, Diego and I rotated the T-240, and put a label on it that N/S is Y, and E/W is X.  Obviously Vert is still Z.

I took a quick look at single arm RIN. Actuating on MC2 vs. the ETM, or using AS55 instead of POY11 made no noticeable difference in the arm cavity RIN.  Not too surprising, but there it is.

Elog from ~5am last night:

Tonight was just several trials of PRFPMI locking, while trying to pay more attention to the lockloss plots each time.

General notes: 

I tried once to acquire DRMI on 1f while the arms were held off resonance.  I wasn't catching lock, so I went back to PRMI+arms.  I aligned the PMC, which I noted in a separate elog.  

I was able to hold the PRMI on REFL33I&Q, and have ALS CARM and DARM at zero CARM offset.  The arm would "buzz" through the resonance regularly.  I use the word buzz because that's kind of what it sounded like.  This is the noise of the ALS system.

I think we want to add the transmission QPD angular signals to the frames.  Right now, we just keep the sums.  It would have been handy to glance at them, and see if they were wiggling in the same way that some other signal was waggling. 

All the data files are in /opt/rtcds/caltech/c1/scripts/LSC/LocklossData.  Each folder is one lockloss.  It includes text files for each trace, as well as any plots that I've made, and any notes taken.  The text files are several MB each, so I'm not going to bog the elog down with them.  There are a few folders that end in "_notInteresting".  These ones are, as one might guess, not interesting.  2 were MC locklosses (I'm not actuating on MC2, so I declared these independent from my work) and one was when I knew that my ALS was bad - the beatnotes weren't in good places, and so the ALS noise was high.

 Folder:  1100342268_POP22goesLow

Working notes:  Lost lock because POP22 went too low.  PRCL and MICH triggered off.  After this, changed PRCL and MICH "down" thresholds to 0.5, from 10.
 

Plots:

Conclusion:  Easy fix.  Changed the down thresholds for MICH and PRCL to be lower, although still low enough that they will trigger off for a true lockloss.  Why though do we lose so much sideband power when the arm transmission goes high?  POP22 dipped below 10 when TRX went above 29.  Does this happen on both sides of the CARM offset?  Quick simulation needed.

------------------------------------------------------------------------

Folder:  1100330534_maybePRCLangular

Working notes:  PRFPMI, reducing CARM offset to arm powers of 7.  CARM on sqrtInv, DARM on DCtrans. PRMI on REFL33 I&Q. Don't know why I lost lock.  Maybe angular stuff in PRC? I think POP spot was moving in yaw as it started to go bad.
Note, later:  regathered data to also get POP angular stuff. Don't think it's POP angular.  Not sure what it is.

Plots:

Conclusion:  I'm not sure what this lockloss was caused by, although it is not something that I can see in the POP QPD (which was my initial suspicion).  It is, like many of the rest of the cases, one where I see significant bounce and roll mode oscillations (error and control signals oscillating at 16 and 24 Hz).  I don't think those are causing the locklosses though.

--------------------------------------------------------------------

Folder:  1100334680_unknown_highArmPowers

Working notes:  PRFPMI, carm_up script finished, sitting at arm powers of 8.  CARM, DARM on DC trans.  PRMI on REFL33.   Don't know why lost lock.

Plots:

[Don't have any? - I'll make some]

Conclusion:  Again, I see 16 and 24 Hz oscillations, but I don't think those are causing the lockloss.

---------------------------------------------------------------------

Folder: 1100331950_unknown

Working notes: PRFPMI, arms about 8.  CARM, DARM on DC trans.  PRMI on REFL33.  Don't know why I lost lock.

Plots:

Conclusion: Don't have an answer.

--------------------------------------------------------

Folder: 1100345981_unknown

Working notes: Lockloss while going to arm powers of 7ish from 6ish.  Not POP angular, POP22 didn't go low.

Plots:

Conclusions:  This one wasn't from POP22 going too low, but again, I don't see anything other than 16 and 24Hz stuff.

I am still staring at / trying to figure out the latter 4 locklosses posted earlier.  But, I have just included the transmission QPD angular output signals to the frames, so we should be able to look at that with locklosses tonight. 

To get the lockloss plots:  in ..../scripts/LSC/LocklossData/ , first run ./FindLockloss.sh <gps time> .  This just pulls the TRX and TRY data, and doesn't save it, so it is pretty quick.  Adjust the gps time until you capture the lockloss in your plot window.  Then run ./LockLossAutoPlot.sh <gps time> to download and save the data.  Since it has become so many channels, it first makes a plot with all of the error and control signals, and then it makes a plot with the power levels and angular signals.  The data folder is just called <gps time>.  I have started also including a text file called notes inside of the folder, with things that I notice in the moment, when I lose lock.  Don't use .txt for the suffix of the notes file, since the ./PlotLockloss.py <folder name> script that will plot data after the fact tries to plot all .txt files.  I have also been appending the folder name with keywords, particularly _notInteresting or _unknown for either obvious lockloss causes or mysterious lockloss cases.

Here is a plot of when the arm powers went pretty high from last night.  CARM and DARM were on ALS comm and diff, PRMI was on REFL33 I&Q.  I set the CARM offset so that I was getting some full arm resonances, and it goes back and forth over the resonance.

The Y axes aren't perfect when I zoom, but the maximum TRX value was 98 in this plot, while the max TRY value was 107. 

MICH_OUT was hitting its digital rails sometimes, and also it looks like PRCL and MICH occasionally lost lock for very brief periods of time. 

Glitch-like events in PRCL_OUT are at the edges of these mini-locklosses. I don't know why POPDC has glitch-y things, but we should see if that's real.

Okay, I've zoomed in a bit, and have found that, interestingly, I see that both POP22 and POP110 decrease, then increase, then decrease again as we pass through full resonance.  This happens in enough places that I'm pretty sure we're not just going back and forth on one side of the resonance, but that we're actually passing through it.  Q pointed out that maybe our demod phase angle is rotating, so I've made some zoom-in plots to see that that's not a significant effect.  I plot the I and Q phases individually, as well as the total=sqrt(I**2 + Q**2), along with TRY (since the increases and decreases are common to both arms, as seen in the plot above).

For POP 22:

For POP 110:

I also plot the MICH and PRCL error signals along with TRY and POP22 total.  You can see that both MICH and PRCL were triggered off about 0.1msec after POP22 total this it's first super low point.  Then, as soon as POP22 becomes large enough, they're triggered back on, which happens about 1.5msec later. (The triggering was actually on POP22I, not POP22tot, but the shapes are the same, and I didn't want to overcrowd my plots).

I am not sure if we consistently lose sideband signal in the PRC more on one side of the CARM resonance than the other, but at least POP22 and POP110 are both lower on the farther side of this particular peak.  I want to think about this more in relation to the simulations that we've done.  One of the more recent things that I see from Q is from September:  elog 10502, although this is looking specifically at the REFL signals at 3f, not the 2f circulating PRCL power as a function of CARM offset.