Summary:

I measured the transfer function of the AO path, and think that there are some features indicative of a problem somewhere in the IMC locking loop.

Details:

Regardless of the locking scheme used, high bandwidth control of the laser frequency relies on the fact that the laser frequency is slaved to the IMC cavity length with nearly zero error below ~50 kHz (assuming the IMC loop has a UGF > 100 kHz). In my single arm experiments, I didn't know what to make of the ripples that became apparent in the measured OLTF as the AO gain was ramped up.

Tonight, I measured the TF of the "AO path", which modifies the error point of the IMC, thereby changing the laser frequency. 

• An SR785 was used to make the measurement.
• The signal was injected at the "EXC B" input on the CM board.
• The CM_SLOW path was disabled, AO gain = 0dB, IMC IN2 gain = 0dB.
• Between "EXC B" and the IMC error point (which I measured at TP1A on the IMC board), we expect that there are 2 poles at ~ 6 Hz, and one pole at ~ 11 Hz.

Attachment #1 shows the result of the measurement. 

• This measurement should be the "Closed Loop Gain" [= 1/(1+L) where L is the open loop gain] of the IMC locking loop. For comparison, I've overlaid the inferred CLG from a measurement of the IMC OLG I made in Jun 2019. The magnitude lines up quite well, but the phase does not 🤔 
• Above 10 kHz, the measurement is as I expect it to be.
• However, between 1 kHz and 10 kHz, I see some periodic features every 1 kHz, which I don't understand. In the IMC OLTF, these would be sharp dips in the OLTF gain.
• I was careful not to overdrive the servo, so I believe these features are not a measurement artefact.
• Combing through past elogs, I couldn't really find any measurements of the IMC OLTF in the 1 kHz - 10 kHz band.
• I decided to measure the spectrum of the IMC error point (with no excitation input), to see if that offered any additional insight. Attachment #2 shows the result - again, periodic features at ~ 1 kHz intervals.

I didn't use POX / POY as a sensor to confirm that this is real frequency noise, I will do so tomorrow. But it may be that realizing a stable crossover is difficult with so many features in the AO path.

Previous thread with a somewhat detailed characterization of the IMC loop electronics.

Tonight was a night of trying to engage the AO path.  The idea was to sit at arm powers of a few on sqrtInvTrans for CARM and ALS for DARM, and try to increase the gain for REFLDC->AO path.

No exciting nit-picky details in locking procedure.  Mostly it was just a night of trying many times. 

The biggest thing that Q and I found tonight was that the 2-pin lemo cable connecting the CM board's SERVO OUT to the MC board's IN2 is shitty.  The symptom that led to this investigation was that I could increase the AO path gain arbitrarily, and have no change in the measured analog CM loop transfer function. We checked that the CM board servo out spit out signals that were roughly what we expected based on our ~2kHz excitation.  However, if we look at digitized signals from the MC board, the noise level was very high, with loads of 60Hz lines, and a teensy-tiny signal peak.  We put a small drive directly into the MC board and could see that, so we determined that the cable is bad.  We have unplugged the white 2-pin lemo, and ran a long BNC cable between the 2 boards.  Tomorrow we need to make a new 2-pin lemo cable so that we can have the lower noise differential drive signal.

After putting in the temporary cable, we do see an excitation sent to the CM board showing up after the MC board.  For this monitoring, the MC_L cable to the ADC has been borrowed, so instead of being the OUT1, the regular length signal, MC_L is currently the OUT2 monitor right after the board inputs. 

At some point in the evening, around 1:15am, ETMX started exhibiting the annoying behavior of wandering off sometimes.  I went in and pushed on the SUS cables to the satellite box, and I think it has helped, although I still saw the drift at least once after the cable-squishing.

Other than that, it has just been many trials.

The best was one where I was holding the arm powers around 4, and got the CM board's AO gain to -8 dB and the MC board's IN2 AO gain to -4 dB. I lost lock trying to increase the CM board gain to -7 dB. 

I took several transfer functions, and used Q's nifty "SRmeasure" script to gather data, and Q made a plot to see the progress.

TF progress plots:

Time series of that lockloss:

I don't know yet if the polarity of the CM board should be plus or minus.  This series was taken with "minus".  But,  since the phase looked opposite of Q's single arm CM board checkout from several months ago, we did a few trials with the polarity switched to "plus".  I thought we weren't getting as high of AO path gains, so I switched back to "minus", but the last few trials didn't get even as far as the plus trials did.  So, I still don't know which sign we want.

More AO efforts. No huge news. 

Came at AO from each side. For each sign, I lost lock just a few dB from the AO portion of the loop crossing unity gain. Both attempts were about arm powers of 1, which should correspond to ~300pm CARM offset, which I have simulated the crossover as possible with my current loop models (including latest MC loop). The gain steps were usually 6dB in between measurements. 

Positive polarity on CM board screen:

I made it to +5 dB of the last plot here, but the 6th broke it open. Gains on CM In2, CM AO, and MC In2 were -6, -4, -2 on that last, lock breaking, step. 

Negative polarity on CM board screen:

Lost it just 2dB above the last trace. Gains were -6, +1, -2 (So, overall 5dB higher than the other polarization)

Many things happened in between these two lock stretches, but I'm not sure what may or may not have affected things. They include:

• Jenne mentioned PRMI being fussy earlier in the evening. I adjusted REFL33 and POP22 angles during a PRMI lock, while CARM was held away with ALS. My simulations suggest that there are small changes to the 3F sensing when the arms are totally absent, but doing it at a finite CARM offset is closer to where we want it, it seems. 
• I tried using REFL165Q for MICH, since my simulations suggest a better MICH/PRCL angle, which would stave off cross couplings. Lined up excitations, etc., but no luck. 
• I measured the PRMI loops
  • found PRCL to have ~200Hz UGF, 8dB gain peaking. Maybe a little high, but didn't seem terrible. 
  • MICH had UGF of around 20Hz, with the FM gain at 0.8. By the shape of the phase bubble, the loop seems designed for higher bandwidth. I raised the gain to 2.5 for a 70ishHz UGF, and called in FMs 7 and 9 for additional triggered boosts. Things seemed to stay locked pretty well. 
• Lower excitation amplitude the second time around, measuring the AO loop. Looking at the CM output spectra, you can see the excitation wailing away; I wanted to avoid it.

The location of the CARM resonance peak lines up with my simulation, which is good, but there appears to be less phase than expected... I tried making sure that we don't have any whitening uncompensated for, but it looked ok. All my AO path loop model contains is the CM board TF (measured and fitted), the IMC seen as an actuator(measured and fitted), and the REFLDC optical TF (simulated in MIST). Maybe the DC path of whatever diode this is coming from needs to be included...

Discontinuities / glitches could be seen in the CM board fast output when MC board gains were changed, which isn't so nice. Incidentally, I notice now that each lock loss corresponded to a step of AO gain on the CM board.

Back in May I looked at all the glitches that happen when we change the AO gain slider on the CM board - see elog 9938.   I wonder if the MC IN2 gain slider has the same issues.  I think I'll look at this this afternoon. Maybe we can set the CM board gain someplace, and just use the MC IN2 slider (if it's not as glitchy) for the delicate part where we're just about to cross unity, and then later we can again use the CM board's AO gain.

EDIT:  Yes, the glitches on the CM board AO path are *much* bigger, and more frequent.  Interestingly, the biggest glitches were every 4 dB.  When I went from -29 to -28, again from -25 to -24, -21 to -20, etc.  I saw the largest glitches on the MC IN2 slider going -29 to -28 and -17 to -16, but if there were small glitches at other transitions, they didn't hit my trigger levels.  I think next time I try engaging the AO path I'll try to do the delicate stuff by upping the MC IN2 gain rather than the CM board AO gain.

[Rana, Jenne]

This evening, we were able to lock the Yarm through the common mode board, using AS55 as our error signal.  Our final UGF is about 5kHz, with 60 degrees of phase margin.

Before dinner, Rana switched the input of the CM board's REFL1 input to be AS55I rather than POY11Q, in the hopes that it would have better SNR.  Demod phase of AS55 was measured to be 14 deg for optimum Yarm->I-phase and has been set to 0 degrees.  Since the POY demod phase had been 90 degrees, which puts in a minus sign, and now we're using 0 deg which doesn't have a minus sign, we're using the plus (instead of minus) polarity of the CM board.

We re-allocated gains to help lower the overall noise by moving 15dB from the CM board AO gain slider to the MC IN2 gain slider, so we weren't attenuating signals.

We see, by taking loop measurements even before engaging the AO path (so, just the digital loop portion) that we've gained something like 20 degrees of phase margin!  We think that about 5 degrees is some LSC loop re-shaping of the boost filter.  We weren't sure why there was a hump of extra gain in the boost filter, so we've created a new (FM8) boost filter which is just a usual resonant gain:  resgain(16.5,7,50)

The cm_down and cm_step scripts in ..../scripts/PRFPMI/ were modified to reflect the settings below, and their current states are included in the tarball attached.

Also, throughout our endeavors this evening, the PC fast rms has stayed nice and low, so we don't suspect any EOM saturation issues.

Now our Yarm digital servo has a gain of -0.0013, with FMs 2, 4, 5, 7, 8 engaged (2, 7, 8 are triggered). 

Our final CM board settings are: 

REFL1 gain = +22dB

offset = -2.898V

Boost = enable

Super Boost = 0

option = disable

1.6k:79 coupled cavity compensator = enabled

polarity = plus

option = disable

AO gain = 15dB

limiter = enable

MC board:  IN1 gain = 18dB, IN2 gain = 0dB.

Here is a measurement of the Common Mode MCL/AO crossover.  The purple/orange trace here is after/before the boost was engaged.

We also have a measurement of the total loop gain, measured with the SR785.  The parameter file, as well as the python script to get the data, are in the tarball attached.  Noteably, the excitation amplitude was 500mV, whereas Q and Rana yesterday were using 5 or 8 mV.  We aren't sure why the big change was necessary to get a reasonable measurement out.  This measurement is with the boost enabled.

Finally, here is a measurement of the MC error point spectra, with the CM boost on, after we reallocated the gains.  There's a giant bump at several tens of kHz.  We need to actually go out with the fast analyzer and tune up the MC loop.

Summary:

1. ​The PRFPMI can be controlled by a mix of ALS and RF signals and circualting arm powers > 100 can be maintained for several tens of minutes at a stretch.
2. The complete RF handoff still cannot be realized - I need to study the AO path crossover more carefully to understand what exactly is wrong and what needs to be done to rectify the problem.

Measurements:

Over the last couple of days, I've been trying to see if I can measure the phase advance due to the AO path - however, I've been unable to do so for any combination of CM board IN1 gain and MC Servo board IN2 gain I've tried. Yesterday, I tried to understand the loop shapes I was measuring a little more, and already, I think I can't explain some features.

Attachment #1 shows the TF measured (using SR785, and the EXC_A bank of the CM board) when the CM Slow path has been engaged.

• All CARM control in this state is digital.
• For the CM Slow path, the digital filter includes a pole at 700 Hz, pole at 5 kHz and zero at 120 Hz (the latter two for coupled cavity pole compensation).
• In this conditions, the arm powers are somewhat stable at ~150, but still there are fluctuations of the order of 50%.
• The "buzzing" as the arms rapidly go in and out of resonance is no longer present though.
• The UGF of the hybrid REFL11+ALS loop is ~200 Hz, with ~45 deg of phase margin.
• Turning off the MC2 violin filters gives some phase back. But I don't really understand the flattening of the TF gain between ~250-500 Hz.

Attachment #2 shows error signal spectra for the in-loop PRFPMI DoFs, for a few different conditions.

• Engaging the REFL11 digital path smooths out the excess noise in the ~30-50 Hz band, which is consistent with the fact that the arm powers stabilize somewhat.
• However, there is some gain peaking around ~400 Hz.
• This is in turn imprinted on the vertex DoFs, making the whole system's stability marginal.

I believe that a stable crossover is hopeless under these conditions.

Next steps:

1. Account for the measured OLTF, understand where the flattening in the few hundred Hz region is coming from.
2. Repeat the high BW POY experiments, but with the simulated coupled cavity pole - maybe this will be a closer simulation to the PRFPMI transition.

There is some evidence of weird saturation but the gain balancing (0.8dB) and orthogonality (~89 deg) for the daughter board on the REFL11 demod board that generates the AO path error signal seem reasonable. This board would probably benefit from the AD797-->Op27 and thick-film-->thin film swap but i don't think this is to blame for being unable to execute the RF transition.

 This afternoon, I labelled both ends of this cable, and reconnected it to the MC servo board. 

[J, Q]

Terse tonight, more verbose tomorrow. 

We have succesfully achieved multiple kHz bandwidth using the CARM AO path. The CM board super boosts are at too high of a frequency to use effectively, given the flattening of the AO TF. 

Jenne's totally, completely, and in all possible ways uncalibrated plot.  Calibration lines are in here (numbers in control room notebook).  I'm going to export and replot the data tomorrow, in real units.

CARM_DARM_AOengaged_23March2015.pdf

For increased flatness of the AO response, and thus less gain peaking in the CARM loop, I reccomend turning down the MC servo VCO gain to 22dB, -6dB of the current setting. 

From there, we should be able to up the overall CARM gain by another 10dB, and turn on a super boost. 

I measured the IN1/IN2 response of the IMC loop with the aglient analyzer providing the IN2 excitation, to see the transfer function of the AO acutation. The hump in the TF explains the flattening out of the CARM OLTF we saw last night. Turning down the gain by 6dB flattens this bump, and more importantly, has around 10dB less gain when the phase goes through -180, meaning more gain margin for the CARM loop. 

Oddly, when I back out the MC OLG from these measurements, the loop shape is different than what Koji and Rana measured in December (ELOG 10841). Specifically, there is some new flattening of the loop shape around 300-400kHz that lowers the frequency where the phase hits -180. What could have caused this???

The -6dB that I mentioned was determined by putting the MC UGF at about 100kHz, at the peak of the phase bubble. This should allow us to safely have a CARM UGF of 40kHz since the MC loop has around +10dB loop gain there, which Rana once quoted as a rule of thumb for these loops. At that UGF, at least one CM board super boost should be fine, based on the loop shapes measured last night. 

Lastly, I also checked out whether the 3 MC super boosts were limiting the AO shape; I did not observe any diffrence of the AO TF when turning off one super boost. It's likely totally fine. 

Jenne has more detailed notes about how things went down last night, but I figure I should write about how we got the AO path stably up. 

As the carm_cm_up script stood after Jenne and Den's work last week, the CARM loop looked like the gold trace in the loop shape plot I posted in the previous elog. The phase bubble was clearly enlarged by the AO path, but there was some bad crossover instability brewing at 400 Hz. This was evident as a large noise peak, and would lead to lock loss if we tried to increase the overall CARM gain.

As with our single arm CM board locking adventures, it was useful to have a filter that made the digital loop shape steeper around the crossover region, so that the 1/f AO+cavity pole shape played nice with the digital slope. As in the single arm trials, this effectively meant undoing the cavity pole compensating zero with a corresponding pole, letting the physical cavity pole do the steepening. This is only possible once the AO path has bestowed some phase upon you. A zero at a somewhat higher frequency (500Hz) gives the digital loop back some phase, which is neccesary to stay locked when the loop has only a few hundred Hz UGF, and the digital phase still matters. This gives us the purple trace. 

This provided us with a loop shape that could smoothly be ramped up in overall gain towards UGFs of multiple kHz (red trace). At this point we could reliably turn on the first boost, which will help in transitioning the PRMI to 1f signals (green trace). We didn't want to ramp it up too much, as we saw that the phase bubble likely ended not much higher than 100kHz, and the OLG magnitude was flattening pretty clearly around 40kHz. While we could turn on a super boost, it didn't look too nice, as we would have to stay at low phase margin to avoid bad gain peaking (blue trace).

As could be seen in the noise spectra that Jenne showed, you can see the violin notches in the CARM noise. This means we are injecting the digital loop noise all over the place. We attempted rolling off the digital loop (by undoing the zero at 500Hz), but found this made the gain at ~200Hz crash down, almost becoming unstable. We likely haven't positioned the crossover frequency in the ideal place for doing this. 

We didn't really give the interferometer any time to see how the long term stability was, since we wanted to poke around and measure as much as we could. While not every attempt would get us all the way there, the current carm_cm_up's success rate at achieving multi-kHz CARM bandwidth was pretty good (probably more than 50%) and the whole thing is still pretty snappy. 

After swapping out the HDD on donatella, I noticed that the display resolution was stuck on 700x400 and could not be changed. To fix this issue, I edited `/etc/apt/sources.list` to include the following:

deb http://ftp.us.debian.org/debian/ testing main non-free contrib
deb-src http://ftp.us.debian.org/debian/ testing main non-free contrib

to make `non-free` packages available in our repository, then I ran:

sudo apt-get update
sudo apt-get install firmware-amd-graphics

After the installation was complete, I did a reboot and the problem was fixed.

The results of the AM/PM measurements:

• Attachment 1: Traces of 9 AM TFs overlaid on top of each other, calibrated by measuring the voltage at the ‘GREEN_REFL’ output where the TF was measured (described in elog 40m:15197). This was almost exactly 2 V.
• Attachment 2: Traces of 9 PM TFs also overlaid measured using DLFD (as described in elog 40m:15180). Calibrated using the measured ~600 mV pk-pk voltage. The phase plots were unwrapped (shifted by 180 deg if needed) so that each started from roughly 0 deg.

Both the AM and PM TFs were scaled to make them have the same average value. Manually adjusting the delay line offset for each measurement using the oscilloscope was probably not accurate enough and therefore resulted in different scaling which this should somewhat compensate.

Attachment 3:

• The orange and green lines are the averages of the PM and AM values of Attachments 1 and 2 respectively.
• The solid red line is at 230 kHz, which was the previously chosen value for PDH locking. The peak seems to have shifted to the left from previous measurements (elog 40m:12077).
• A horizontal black dashed line is drawn to show where the ratio is 10^5.
• The red regions correspond to frequencies where PM/AM > 10^5 [only shown for frequencies greater than 200kHz], these are roughly (in kHz):
  • 211.4-213.9
  • 221.4-230.7 (peak at 225.642)
  • 240.8-257.9
  • ~748.3
  • 753.3-799.8, two largest peaks at 763.673 and 770.237
  • 809.6-829.3, peak at 819.472
  • 839.2-842.4
  • 881.8-891.7

Updated Calibration

The new plot is in Attachment 5.

The new calibration factor used: 5 MHz/V at the output of the mixer to obtain the frequency modulation and then division by the mod. freq. to obtain PM.

5 MHz/V because changing the PZT voltage by 0.01 V=> change in beat frequency by 0.1 MHz, which was seen as a 20 mV change in the delay line mixer output.

Again, the calibration is not very precise and I will probably repeat this experiment at some point more precisely.

The AM transfer function of the Y end laser has been measured again, but using the frequency-doubled laser this time.

Here is the latest plot of the AM transfer function. The Y-axis is calibrated to RIN (Relative Intensity Noise) / V.

IFBW (which corresponds to a frequency resolution) was set to 100 Hz and the data was averaged about 40 times in a frequency range of 100 kHz - 400 kHz.

Also the zipped data is attached.

It is obvious that out current modulation frequency of 179 kHz (178850 Hz) is not at any of the notches.

It could potentially introduce some amount of the offset to the PDH signal, which allows the audio frequency AM noise to couple into the PDH signal.

Currently I am measuring how much offset we have had because of the mismatched modulation frequency and how much the offset can be reduced by tuning the modulation frequency.

 Keiko, Anamaria

We started to investigate the AM modulation mistery again. Checking just after the EOM, there are AM modulation about -45dBm. Even if we adjust the HWP just before the EOM, AM components grow up in 5 mins. This is the same situation as before. Only the difference from before is that we don't have PBS and HWP between the EOM and the monitor PD. So we have a simpler setup this time.

We will try to align the pockells cell alignment tomorrow daytime, as it may be a problem when the crystal and the beam are not well parallel. This adjustment has been done before and it didn't improve AM level at that time.

Keiko, Suresh

AM modulations are still there ... the mechanical design for the stages, RF cables, and connections are not good and affecting the alignment.

I write the activity in the time series this time - Because we suspect the slight EOM misalignment to the beam produces the unwanted AM sidebands, we tried to align the EOM as much as possible. First I aligned the EOM tilt aligner so that the maximum power goes through. I found that about 5% power was dumped by EOM. After adjusting the alignment, the AM modulation seemed be much better and stable, however, it came up after about 20 mins. They grew up up to about -40dBm, while the noise floor is -60 dBm (when AM is minimised, with DC power of 8V by PDA225 photodetector).

We changed the EOM stage (below the tilt aligner) from a small plate to a large plate, so that the EOM base can be more stable. The EOM stands on the pile of several black plate. There was a gap below the tilt aligner because of a small plate.  So we swapped the small plate to large plate to eliminate the springly gap. However it didn't make any difference - it is the current status and there is still AM modulations right now.

During above activities, we leaned that the main cause of the EOM misalignment may be the RF cables and the resonator box connected to the EOM. They are connected to the EOM by an SMA adaptor, not any soft cables. It is very likely applying some  torc force to the EOM box. The resonator box is almost hunging from the EOM case and just your slight touch changes EOM alinment quite a bit and AM mod becomes large. 

I will replace the SMA connector between the resonator box and EOM to be a soft cable, so that the box doesn't hung from EOM tomorrow. Also, I will measure the AM mod depth so that we compare with the PM mod depth.

AM modulation depths are found to be 50 times smaller than PM modulation depths.

m(AM,f1) ~ m(AM, f2) = 0.003 while m(PM, f1)=0.17 and m(PM, f2)=0.19.

Measured values;

* DC power = 5.2V which is assumed to be 0.74mW according to the PDA255 manual.

*AM_f1 and AM_f2 power = -55.9 dBm = 2.5 * 10^(-9) W.

AM f2 power is assumed to be the similar value of f1. I can't measure f2 (55MHz) level properly because the PD (PDA255) is 50MHz bandwidth. From the (P_SB/P_CR) = (m/2) ^2 relation where P_SB and P_CR are the sideband and carrier power, respectively, I estimated the rough the AM modulation depths. Although DC power include the AM SB powers, I assumed that SB powers are enough small and the DC power can be considered as the carrier power, P_CR. The resulting modulation depth is about 0.003.

On the other hand, from the OSA, today's PM mod depths are 0.17 and 0.19 for f1 and f2, respectively. Please note that these numbers contains (small) AM sidebands components too. Comparing with the PM and AM sideband depths, AM sidebands seems to be enough small.

AM modulation will add offset on SRCL signal as well as PRCL signal. About 2% of the signal amplitude with the current AM level. MICH will not be affected very much.

From #5504, as for the AM modulation I checked the MICH and SRCL signals in addition to the last post for PRCL, to see the AM modulation effect on those signals. On the last post, PRCL (REFL11I) was found to have 0.002 while the maximum signal amplitude is 0.15 we use . Here, I did the same simulation for MICH and SRCL.

As a result, MICH signals are not affected very much. The AM modulation slightly changes signal slopes, but doesn't add offsets apparently. SRCL is affected more, for REFL signals. All the REFL channels get about 0.0015 offsets while the signal ampliture varies up to 0.002. AS55I (currently used for SRCL) has 1e-7 offset for 6e-6 amplitude signal (in the last figure) - which is the same offset ratio comparing with the amplitude in the PRCL case -

(1) MICH signals at AS port with AM m=0

(2) MICH signals at AS port with AM m=0.003

(3) SRCL signals at AS/REFL port with AM m=0

(3) SRCL signals at AS/REFL port with AM m=0.003

From the night day before yesterday (Sep 22nd, Thursday night. Sorry for my late update), there are more AM modulations than I measured in the previous post. It is changing a lot, indeed! Looking at the REFL11 I and Q signals on the dataviewer, the signal offset were huge, even after "LSCoffset" script. Probably the modulation index of AM was same order of PM at that time. The level of AM mod index is changing a lot depending on the EOM alingment which is not very stable, and also on the environment such as temperature .

To reduce AM modulations, here I note some suggestions you may want to try :

* Change the SAM connectors between RF resonator and EOM to be a soft but short connector, so that the resonator box doesn't hung from the EOM.

* Change the RF resonator base to be stable posts. Now several black plates are piled to make one base.

* Install a temperature shield

* Also probably you want to change the BNC connector on the RF resonator to be SMA.

* Be careful of the EOM yaw alignment. Pitch seemed to be less sensitive in producing AM than yaw alignment.

I'd like to see some details about how to determine that the ratio of 1:50 is small enough for AM:PM.

* What have people achieved in past according to the elogs©  of the measurements?

* What do we expect the effect of 1:50 to be? How much offset does this make in the MICH/PRC/SRC loops? How much offset is too much?

Recall that we are using frontal modulation with a rather small Schnupp Asymmetry...

The signal offset due to the AM modulation is estimated by a simulation for PRCL for now. Please see the result below.

Too see how bad or good the AM modulation with 1/50 modulation depths of PM, I ran a simulation. For example I looked at PRCL sweep signal for each channel. I tried the three AM modulation depths, (1) m_AM=0 & m_PM = 0.17 (2) m_AM = 0.003 & m_PM = 0.17 which is the current modulation situation (3) m_AM = 0.17 & m_PM = 0.17 in which AM is the same modulation depth as PM.  For the current status of (2), there are offsets on signals up to 0.002 while the maximum signal amplitude is 0.15. I can't tell how bad it is.... Any suggestions?

(1) m_AM=0 & m_PM = 0.17. There is no offset in the signals.

(2) m_AM = 0.003 & m_PM = 0.17. There are offsets on signals up to 0.002 while the maximum signal amplitude is 0.15.

(3) m_AM = 0.17 & m_PM = 0.17. There are offsets on signals up to 0.1 while the maximum signal amplitude is 0.2.

I will look at MICH and SRCL in the same way. 

How about changing the x-axis of all these plots into meters or picometers and tell us how wide the PRC resonance is? (something similar to the arm cavity linewidth expression)

Also, there's the question of the relative AM/PM phase. I think you have to try out both I & Q in the sim. I think we expect Q to be the most effected by AM.

[Kiwamu, Mirko]

Non-optimal 11MHz SB frequency causes PM to be transformed into AM.
m_AM / m_PM = 4039 * 1kHz / df , with df beeing the amount the SB freq. is off.

Someone might want to double check ths.

 Actually there was an error.

For 11MHz it is:
m_AM / m_PM = 2228 * 1kHz / df

For 55MHz:
m_AM / m_PM = 99.80 * 1kHz / df

see PDF

Big Johnny and I hacked a function generator output into the cross-connect of the 80 MHz VCO driver so that we could modulate the

amplitude of the light going into the RefCav. The goal of this is to measure the coefficient between cavity power fluctuations and the

apparent length fluctuations. This is to see if the thermo-optic noise in coatings behaves like we expect.

To do this we disconnected the wire #2 (white wire) at the cross-connect for the 9-pin D-sub which powers the VCO driver. This is

called VCOMODLEVEL (on the schematic and the screen). In the box, this modulates the gain in the homemade high power Amp which

sends the actual VCO signal to the AOM.

This signal is filtered inside the box by 2 poles at 34 Hz. I injected a sine wave of 3 Vpp into this input. The mean value was 4.6 V. The

RCTRANSPD = 0.83 Vdc. We measure a a peak there of 1.5 mVrms. To measure the frequency peak we look in

the FSS_FAST signal from the VME interface card. With a 10 mHz linewidth, there's no peak in the data above the background. This signal

is basically a direct measure of the signal going to the NPRO PZT, so the calibration is 1.1 MHz/V.

We expect a coefficient of ~20 Hz/uW (input power fluctuations). We have ~1 mW into the RC, so we might expect a ~20 Hz frequency shift.

That would be a peak-height of 20 uV. In fact, we get an upper limit of 10 uV.

 Later, with more averaging, we get an upper limit of 1e-3 V/V which translates to 1e-3 * 1.1 MHz / 1 mW ~ 1 Hz/uW. This is substantially lower

than the numbers in most of the frequency stabilization papers. Perhaps, this cavity has a very low absorption?

[Shruti, Rana]

- At the X end, we set up the network analyzer to begin measurement of the AM transfer function by actuation of the laser PZT.

- The lid of the PDH optics setup was removed to make some checks and then replaced.

- From the PDH servo electronics setup the 'GREEN_REFL' and 'TO AUX-X LASER PZT' cables were removed for the measurement and then re-attached after.

- The signal today was too low to make a real measurement of the AM transfer function, but the GPIB scripts and interfacing was tested. 

Some details:

- There was a SR560+SR785 (not connected for measurement) placed near the X end which I moved; it is now behind the electronics rack by the X arm beam tube (~15m away).

- Also, for the AM measurement I moved the AG5395A from behind the PSL setup to the X end, where it now is.

- By toggling the XGREEN shutter, I noticed that the cavity was not resonant before I disconnected anything from the setup since the spot shape kept changing, but I proceeded anyway. 

- Because Rana said that it was important for me to mention: the ~5 USD blue-yellow crocs (that I now use) work fine for me.

The AM Measurement:

1. The cables were calibrated with the DC block in the A port (for a A/R measurement)

2. The cable to the PZT was disconnected from the pomona box and connected to the RF out of the NA, the PD output labelled 'GREEN_REFL' was also disconnected and connected to the B port via a DC block. 

3. The ITMX was 'misaligned'. (This allowed the reflected green PD output as seen on the oscilloscope to stabilize.)

4. The PZT is modulated in frequency and the residual amplitude modulation (as observed in the measured reflected green light) is plotted, ref. Attachment 1. The parameters for the plotted data in the attachment were:

# AG4395A Measurement - Timestamp: Nov 07 2019 - 17:04:07
#---------- Measurement Parameters ------------
# Start Frequency (Hz): 10000.0, 10000.0
# Stop Frequency (Hz): 10000000.0, 10000000.0
# Frequency Points: 801, 801
# Measurement Format: LOGM, PHAS
# Measuremed Input: AR, AR
#---------- Analyzer Settings ----------
# Number of Averages: 8
# Auto Bandwidth: On, On
# IF Bandwidth: 300.0, 300.0
# Input Attenuators (R,A,B): 0dB 10dB 20dB 
# Excitation amplitude = -10.0dBm

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

Update (19:13 7thNov19):  When the ITMX was intentionally misaligned, Rana and I checked to see if the Oplevs were turned off and they were. But while I was casually checking the Oplevs again, they were on! 

Not sure what to do about this or what caused it. 

Kiwamu, Keiko, Anamaria

Looking at the I and Q signals coming from REFL11 and REFL55 we saw large offsets, which would mean we have amplitude modulation, especially at 11MHz. We checked the PD themselves with RF spectrum analyzer, and at their frequencies we see stationary peaks (even if we look only at direct reflection from PRM). We changed the attenuation of the PSL EOM, and saw the peak go down. So first check is beam out of PSL EOM, to make sure the input beam is aligned to the crystal axis and is not giving AM modulation in adition to PM.

I took a few AM TF measurements at the X end for which I:

• Misaligned the ITMX (then re-aligned it)
• Opened the X green shutter during the measurements and closed it at the end
• Moved the Agilent from the PSL area to the X end, the delay line and mixer still remains near the PSL area (will move it soon)
• Took a bunch of TFs

I will post the data soon.

[Mirko / Kiwamu]

 We have reviewed the AM issue and confirmed the ratio of AM vs. PM had been about 6 x10³.

The ratio sounds reasonably big, but in reality we still have some amount of offsets in the LSC demod signals.

Next week, Mirko will estimate the effect from a mismatch in the MC absolute length and the modulation frequency.

(Details)

 Please correct us if something is wrong in the calculations.

 According to the measurement done by Keiko (#5502):

        DC = 5.2 V

        AM @ 11 and 55 MHz = - 56 dBm = 0.35 mV (in 50 Ohm system)

Therefore the intensity modulation is 0.35 mV / 5.2 V = 6.7 x 10^-5

Since the AM index is half of the intensity modulation index, our AM index is now about 3.4 x 10^-5

According to Mirko's OSA measurement, the PM index have been about 0.2.

As a result,  PM/AM = 6 x 10³

[Koji, Suresh]

In the previous measurement, the PDA 255 had most probably saturated at DC, since the maximum ouput voltage of PDA255 is 5V when it is driving a 50 Ohm load.  It has a bandwidth of 0 to 50MHz and so can be reliably used to measure only the 11 MHz AM peak.  In this band it has a conversion efficiency of 7000 V per Watt (optical power at 1064nm).  [Conversion efficiency:  From the data sheet we get 0.7 A/W of photo-current at 1064nm and 10^4 V/A of transimpedance]  The transimpedance at 55 MHz is not given in the data sheet.  Even if PDA255 is driving a high impedance load, at high incident power levels the bandwidth will be reduced due to finite gain x bandwidth product of the opamps involved, so the conversion efficiency at 11 MHz would not be equal to that at DC.

So Koji repeated the measurement with a lower incident light level:

**********************************

V_DC = 1.07 V  with 50 Ohm termination on the multimeter.

Peak height at 11 MHz on the spectrum analyzer (50 Ohm input termination) = -48.54 dBm

***********************************

Calculation: 

a) RF_Power at 11 MHz :  -48.45 dBm = 1.4 x 10^(-8) W

b) RF_Power = [(V_rms)^2] / 50_ohm  ==> V_rms = 8.4 x 10^(-4) V

c) Optical Power at 11 MHz: [V_rms / 7000] = 1.2 x 10^(-7) W

d) Optical Power at DC =  [V_DC / 7000] = 1.46 x 10^(-4) W

e) Intensity ratio:  I_AM / I_c = 7.9 x 10^(-4) . AM:Carrier amplitude ratio is half of the intensity ratio = 4.0 x 10^(-4)

f) PM amplitude ratio from Mirko's measurement is 0.2

g) The PM to AM amplitude ratio is 506

_________________________________

As the AM peak is highly dependent upon the drifting EOM position in yaw, it is quite likely that a higher PM/AM ratio could occur.  But this measurement shows how small it could get if the current situation is allowed to continue.

Correction: Koji noted that Mirko actually reports a PM modulation index of 0.17 for the 11 MHz sideband (elog: http://nodus.ligo.caltech.edu:8080/40m/5462. This means

f) the amplitude ratio of the PM side-band to carrier is half of that = 0.084

g)  the PM to AM amplitude ratio as 0.084 / [4.0 x 10^(-4)]  = 209.

The end sus models (c1scx and c1scy) both contain some ALS stuff.  This stuff could maybe be moved to their own models, but whatever.

The stuff at X and Y were identical, but were code copies (BAD!).  I made a new library part for the ALS end controls: ${userapps}/isc/c1/model/ALS_END.mdl

It contains just some filter modules for the ALS end laser control, and a monitor of the ALS end REFL PD DC.  I also added a DQ block for the recorded channels (see screen shot).

When I added this new part to c1scx and c1scy I made it so the channel names would be more sensible.  Instead of "GCX" and "GCY", they are now "ALS-X" and "ALS-Y".  They will now all show up under the ALS subsystem.

This is a good modification. We just need to check how the ALS scripts are affected.

[ericq, gautam]

Gautam couldn't observe a Y green beatnote earlier, so we checked things out, fixed things up, and performance is back to nominal based on past references. 

Things done:

• Marconi carrier output switched back on after Koji's excellent RF maintence
• BBPD power supplies switched on
• Removed a steering mirror from the green beatY path to do near/far field alignment. 
• Aligned PSL / Y green beams 
• Replaced mirror, centered beam on BBPD, moved GTRY camera to get the new spot.
• POY locked, dither aligned, beatnote found, checked ALS out-of-loop noise, found to be in good shape. 

Sorry, I completely forgot to turn the Marconi on...

[ericq, Gautam]

We checked the UGF of the AUX X PDH servo, found a ~6kHz UGF with ~45 degree phase margin, with the gain dial maxed out at 10.0. Laser current is at 1.90, direct IR output is ~300mW.

We recovered ALS readout of IR-locked arms. While the GTRX seemed low, after touching up the beam alignment, the DFD was reporting a healthy amount of signal. ALSY was perfectly nominal. 

ALSX was a good deal higher than usual. Furthermore, there's a weird shape around ~1kHz that I can't explain at this point. It's present in both the IR and green beats. I don't suspect the DFD electronics, because the Y beat came through fine. The peak has moderate coherence with the AUX X PDH error signal (0.5 or so), but the shape of the PDH error signal is mostly smooth in the band in which the phase tracker output is wonky, but a hint of the bump is present. 

Turning the PDH loop gain down increases the power spectrum of the error signal, obviously, but also smoothens out the phase tracker output. The PDH error signal spectrum in the G=10 case via DTT is drowning in ADC noise a bit, so we grabbed it's spectrum with the SR785 (attachment #2, ASD in V/rtHz), to show the smoothness thereof.

Finally, we took the X PDH box to the Y end to see how ALSY would perform, to see if the box was to blame. Right off the bat, when examining the spectrum of error signal with the X box, we see many large peaks in the tens of kHz, which are not present at the same gain with the Y PDH box. Some opamp oscillation shenanigans may be afoot... BUUUUUT: when swapping the Y PDH box into the X PDH setup, the ~1kHz bump is identical. ugh

While carrying out my end-table power investigations, I decided to take a quick look at the out-of-loop ALSX noise - see the attached plot. The feature at ~1kHz seems less prominent (factor of 2?) now, though its still present, and the overall noise above a few tens of Hz is still much higher than the reference. The green transmission was maximized to ~0.19 before this spectrum was taken.

EDIT 1130pm: 

We managed to access the trends for the green reflected and transmitted powers from a couple of months back when things were in their nominal state - see Attachment #2 for the situation then. For the X arm, the green reflected power has gone down from ~1300 counts (November 2015) to ~600 counts (january 2016) when locked to the arm and alignment is optimized. The corresponding numbers for the green transmitted powers (PSL + End Laser) are 0.47 (November 2015) and ~0.18 (January 2016). This seems to be a pretty dramatic change over just two months. For the Y-arm, the numbers are: ~3500 counts (Green REFL, Nov 2015), ~3500 counts (Green REFL, Jan 2016) ~1.3 (Green Trans, Nov 2015), ~1 (Green Trans, Jan 2016). So it definitely looks like something has changed dramatically with the X-end setup, while the Y-end seems consistent with what we had a couple of months ago...

I have been looking at the X-end ALS setup.
I was playing with the control bandwidth to see the effect to the phase tracker output (i.e. ALS err).
For this test the arm was locked with the IR and the green beat note was used as the monitor.

From the shape of the error signal, the UGF of the green PDH was ~10kHz. When I increased the gain
to make the servo peaky, actually the floor level of the ALS err became WORSE. I did not see any improvement
anywhere. So, high residual error RMS cause some broadband noise in the ALS??? This should be checked.
Then when the UGF was lowered to 3kHz, I could see some bump at 3kHz showed up in the ALS error.
I didn't see the change of the PSD below 1kHz. So, more supression of the green PDH does not help
to improve the ALS error?

Then, I started to play with the phase tracker. It seems that someone already added the LF booster
to the phase tracker servo. I checked the phase tracker error  and confirmed it is well supressed.
Further integrator does not help to reduce the phase tracker error.

For the next thing I started to change the offset of the phase tracker. This actually changes
the ALS error level! The attached plot shows the dependence of the ALS error PSD on the phase tracker
output. At the time of this measurement, the offset of -10 exhibited the best noise level.
This was, indeed, factor of 3~5 improvement compared to the zero offset case below 100Hz.

I'm afraid that this offset changes the beat frequency as I had the best noise level at the offset of -5
with a different lock streatch. We should look at this more carefully. If the beat freq changes the offset,
this give us another reason to fix the beat frequency (i.e. we need the frequency control loop.

= Today's ALSX error would have not been the usual low noise state.
We should recover the nominal state of the ALS and make the same test =

 After working some more on the EY table, we are getting some TEM00 flashes for the Y arm green. We have had to raise the height of one of the MM lenses to prevent clipping.

We used a function generator to apply a ~300 mV 10 Hz triangle wave to scan the laser frequency while aligning.

We tried to use the C1:ALS-TRY_OUT channel to help us in our alignment but there are a couple problems:

1) It seems that there is an uncompensated whitening filter before the ADC - Annalisa is making a compensation filter now.

2) The data delay is too much to use this for fast alignment. We might need to get a coax cable down there or mount a wired ethernet computer on the wall.

3) We need to make DQ channels for the TRY and TRX OUT. We need long term data of these, not just test points.

 I made the anti-whitening filter for the C1:ALS-TRY_OUT channel. But then I forgot to make an ELOG because I am bad.

 [Jenne, Annalisa]

DQ channels have been created in the C1ALS model for TRX and TRY. They are called TRX_OUT and TRY_OUT and the sampling rate is 2048 Hz.

I've switches the ALS, ASC, and LSC plots on the summary pages from plotting raw frames, to plotting minute trends, instead. Now, the plots contain information, instead of being completely blank, but data is not recorded on the plots after 12UTC.

Typically, I make changes to the summary pages on my own version of the pages, found at https://ldas-jobs.ligo.caltech.edu/~eve.chase/summary/day/, where I change the summary pages for June 30 and then import such changes into the main summary pages. 

[ericQ, gautam]

Today we spent some time looking into the PDH situation at the X end. A summary of our findings.

1. There is something that I don't understand with regards to the modulation signal being sent to the laser PZT via the sum+HPF pomona box - it used to be that with 2Vpp signal from the function generator, we got ~5mVpp signal at the PZT, which with the old specs resulted in a modulation of ~0.12rad. Now, however, I found that there was a need to place a 20dB attenuator after the splitter from the function generator in order to realize a modulation depth of ~0.25 (which is what we aim for, measured by locking to the TEM00 modes of the carrier and sidebands and comparing the ratio of powers). It could be that the PZT capacitance has changed dramatically after the repair. Nevertheless, I still cant reconcile the numbers. We measured the transfer function from the LO input of the pomona box to the output with the PZT connected, and figure there should be ~70dB of attentuation (with the 20dB additional attenuator in place). But this means 1Vpp*0.0003*70rad/V = 0.02rad which is an order of magnitude away from what the ratio of powers suggest. Maybe the measurement technique was not valid. In any case, this setup appears to work, and I'm also able to send +7dBm to the mixer which is what it wants (function generator output is 3Vpp).
2. In addition to the above, I found that the demodulated error signal had a peak-to-peak of a few volts. But the PDH servo is designed to have tens of mV at the input. Hence, it was necessary to turn down the gain of the REFL PD to 10dB and add a 20dB attenuator between mixer output and servo input.
3. While Johannes and I were investigating this earlier in the afternoon, we found that the waveform going to the laser PZT was weirdly distorted (still kind of sinusoidal in shape, but more rounded, I will put up a picture shortly). This may not be the biggest problem, but perhaps there is a better way to pipe the LO signal to the PZT and mixer than what is currently done.
4. We then looked at loop transfer function and spectrum of the control signal. Plots to follow. They look okay.
5. I measured the green power coming onto the PSL table. It is ~400uW. After optimizing alignment, the green transmission is ~0.4 according to whatever old normalization we are using.
6. We then recovered the X green beatnote and looked at the ALS noise spectrum. Beatnote amplitude at the beat PD is ~ -27dBm. The coherence in the region of a few hundred Hz suggests that some improvements can be made to the PDH situation (the gain of the PDH servo is maxed out at the X end at the moment...). But the bottom line is this is probably good enough to get back to locking...

All seems very fishy. Its not good to put attenuators and filters in nilly-willy.

1. Once the post-PD bandpass has been designed and constructed, you should be able to use whatever PD gain setting gives you the best SNR. There's no need to use more PD gain than necessary; it just reduces the PD bandwidth. What is the input referred current noise of the PD at the different gain settings?
2. The open loop mixer output *should* be very large. It should be reduced to mV only when the loop is closed.
3. The better way to estimate the modulation depth is to lock the arm on red as usual and then scan the EX laser and look at the green transmission. The FSR is 3.7 MHz, so the SBs should show up well in a narrow scan around the carrier.
4. I guess its going to be tough to impedance match the splitter box to the NPRO PZT, since its impedance is all over the place at 200-300 kHz, but you could put a 50 Ohm in-line terminator in there somewhere?
5. The Bode plot seems to indicate that we could easily get a 10 kHz UGF and then switch on a Boost. Is the remote Boost switch disabled or always ON? I am suspicious of the plot and think that the coarse trace is probably missing some sharp resonances which will sneakily bite you.

After Jenne and Masayuki told that they were not able to stabilize the ALS for either arms yesterday, I looked into things with the ALS servo.

I had trouble initially trying to even stabilize the loop for a few minutes. So I measured the OLTF of the phase tracker loop and the ALS X arm servo. I changed phase tracker gain to 125 and that rendered UGF of 2KHz and phase margin of 45 degrees for the phase tracker loop.

The ALS servo gain was set such that UGF was 125Hz and phase margin 38 degrees (attached is the transfer function measurement for the servo).

I could stabilize the arm to ~500 Hz/rtHz (rms), which is twice that of what we had while we did the (PRMI+1arm ALS).

But ALS was still not stable long enough with the higher rms to even allow a cavity scan to find IR resonance. I suspect the problem to now lie with the PDH loop. We should be looking to stabilize the PDH for green if we need a stable ALS.

I'm running a test to see how stable the EX green lock is. For this purpose, I've left the slow temperature tuning servo on (there is a 100 count limiter enabled, so nothing crazy should happen).

- ALS X/Y arm stability was checked by IR locked arms.

- Basically the stability looks same as before.

Q sez: here are some ALS ASDs (in Hz/rtHz). 

The reference plots are with the arms locked on CARM/DARM with ALS. The main traces are with the arms locked on POX/POY. Alignment affects these traces a fair amount.

The X arm ALS seems no worse for the upgrade, and the PZT actuators do look pretty orthogonal when we play around with the alignment. 

While setting up for this measurement, I noticed something odd with the whitening switching for the ALS channels. For the usual LSC channels, the whitening is set up such that switching FM1 on the MEDM screen changes a BIO bit which then enables/disables the analog whitening stage. But this feature doesn't seem to be working for the ALS channels - I terminated all 4 channels at the LSC rack, and measured the spectrum of the IN1 signals with DTT in the two settings, such that I expect to see a difference in the spectra if the whitening is enabled or disabled - FM1 enabled (expected analog whitening to be engaged) and FM1 disabled (expected analog whitening to be bypassed). But I see no difference in the spectra. I confirmed that the BIO bit switching is happening at least on the software level (i.e. the bit indicator MEDM screens indicate state toggling when FM1 is ON/OFF). But I don't know if something is amiss in the signal chain, especially since we are using Hardware channels that were previously used for AS_165 and POP_55 signals.

Is the whitening shape such that we expect the terminated noise level to be below ADC Noise even when the whitening is engaged? I just checked the shape of the de-whitening filter, and it has -40dB gain above 150Hz, so the inverse shape should have +40dB gain. 

gautam 2.15pm: This was a FALSE ALARM, with the inputs terminated, the electronics noise really is that low such that it is buried under ADC noise even with +40dB gain. I cranked up the flat whitening gain from 0dB to 45dB for the X channels (but left the Y channels at 0dB). Attachment #2 is the comparison. Looks like the switching works just fine.