Wonkey shape: Looks like a loop supression. Your http://nodus.ligo.caltech.edu:8080/40m/11016 also suggests it too, doesn't it?

Dang it, yes. You're right.  I should have caught that. 

Since Dcpl and Srefl are both zero during the measurement (since it was an ALS lock), this is actually

So, I need to remove the effect of the ALS closed loop, to get the actual quantity I was looking for.

I re-did the Mathematica notebook according to the most current diagram (note to daytime self: attach .nb file!!!), and found that the denominator has changed, such that plugging in the new D=-A_refl*P_als*S_als gives the same

full-system closed loop gain of   

where   is the open loop gain, and the * indicates either the REFL or ALS portions of the system. 

I have also plotted some things with Matlab, although I'm a little confused, and my daytime self needs to spend some more time thinking about this.

In the actuators (both for REFL and ALS), I include a pendulum, the digital anti-imaging filters that let us go from the 16kHz model to the 64kHz IOP and the analog anti-imaging filters after the DAC.  Note to self:  still need to include violin filters here.

For the servo gains, I copy the filters that we are using from Foton, and give them the same overall gain multiplier that is in the filter bank.  For the ALS going through the CARM filter bank, this is FMs 1, 2, 3, 5, 6 with a gain of 15.  For the RF (actually, POY here) going through the MC filter bank, this is FMs 4, 5, 7 with a gain of 0.08. 

For the plants of each system, since this is still single arm lock, I just include a cavity pole (80kHz for ALS, 18kHz for REFL). 

In the sensors (both for REFL and ALS), I include the analog anti-aliasing as well as the digital anti-aliasing to allow us to go from the 64kHz IOP to the 16kHz front end system.  For the ALS I also include in the sensor the closed loop response of the phase tracker loop (H/(1-H), where H is the open loop gain of the phase tracker).  For both sensors, I also include a semi-arbitrary number to make the full single-loop open loop gain have a UGF of 200Hz.  In the ALS sensor, I also include a minus sign to make the full open loop gain have the correct phase.

Here I plot the open loop gains of the individual single loops, as well as the open loop gain of the full system (Hals + Hrefl - Hals*Hrefl).  I feel like I must be missing a minus sign in my ALS loop, but I don't know where, and my nighttime brain doesn't want to just throw in minus signs without knowing why.  That will affect how the final ALSfool (blue trace) looks, so maybe it's not really as crazy as it looks right now.

Also, I was trying to explain to myself why we are getting the shape that we are in our measurements of the cancellation (http://nodus.ligo.caltech.edu:8080/40m/11041).  But, I can't.  Below are the plots of the transfer functions from either point 9 or 10 (before or after the G_refl) to point 5, which is the ALS error point.  The measurement in elog 11041 should correspond to the blue trace here.  For these traces, the decoupling is set to just (-A_refl), although there aren't any noticeable changes in the shape if I just set D=0.  If we start with the assumption that D=0, the shape and magnitude are basically identical to this plot, and then as we make D=-A_refl P_als S_als, the transfer functions both go to zero. 

So.  Why is it that with no decoupling, the transfer function from 10 to 5 is tiny?  Why do the shapes plotted below look nothing at all like the measured cancellation shape?  Daytime brain needs to think some more.

Here is an updated cartoon, with the ALS sensor explicitly shown as the beatbox times the closed loop response of the phase tracker servo. 

The most important transfer functions are written on the diagram.  Others can be extracted from the attached Mathematica file (which corresponds to this diagram).

The ALS fool scheme is now diagrammed up in OmniGraffle, including its new official icon.  The mathematica notebook has not yet been updated.

EDIT, JCD, 17Feb2015:  Updated cartoon and calculation: http://131.215.115.52:8080/40m/11043

[Jenne, EricQ]

We tried several times tonight to engage the Fool path with the PRFPMI.  No success. 

First, we locked the arms on ALS, in CARM/DARM mode, and measured the cancellation ability, to make sure that the filter shape and gain was set correctly.  For the PRFPMI, it was okay using the same shape as the single arm case, but the gain was +20.0.  There might be a bit more cancellation to be gained if we adjust the shape at the ~1dB level, but we're already able to get 20dB of cancellation, so we decided that would be enough to give things a try.  To get this much cancellation, we set the phase tracker loops to both have 2kHz UGFs, almost exactly.  We should implement a UGF servo, or the amplitude method version of that as Koji suggested ages ago, so that the phase tracker is always at the same place.

I don't think that REFL 11 is seeing as much CARM as I expect.  We ended up switching over to linearized REFL55 for our attempts.  When we're close to zero CARM offset, the arms are constantly flashing through resonance, and we get the loud buzzing.  REFL11 doesn't seem to see any of this, even though we should be close enough to see some PDH action.  REFL55 does change as we get closer to resonance, so I think it's seeing some real CARM stuff.

We tried engaging the Fool, but I don't think it did anything too useful. We need to make an estimate of what we expect our gain of the REFL loop to be - or at least the sign.

The PRMI is still not stable enough.  It keeps falling out of lock when we get to high-ish arm powers.  Not good.  More brain power tomorrow on the modulation cancellation issue. 

Perhaps if things are stable at moderate arm powers, we can use an excitation to line up the ALS vs. REFL error signals, and then watch the noises of them change as we move around in CARM offset.  This should tell us when the linearized REFL signal is quiet enough that it's worth triggering and trying to transfer over.

The last lockloss tonight, there was something funny going on, that we can't explain.  Even though both arms were locked on the CARM/DARM combined ALS signals, beatx doesn't see the giant oscillation that causes carm to lose lock until much later.  Fool was trying to do something, but that should affect both als individal signals in the same way.  Mystery.

Leaving a note on the ALS feedback before I forget:

The MC2 suspension needs to have an input for the ALS feedback in the realtime model like ETMs.

I added an ALS feedback path on the MC2 suspension and this path will enable us to stabilise the MC length using the ALS scheme.

[Koji, gautam]

Summary:

We think we got to the bottom of this issue today. The RF signal level going into the demod board is too high. This electronics chain needs some careful gain reallocation.

Details:

I was demonstrating to Koji a strange feature I had noticed in the ALS control, whereby when applying a CARM offset to detune the arms, the two arms seemed to respond differently (based on the transmission levels). This kind of CARM-->DARM coupling seemed strange to me. Anyway, I also noticed that the EPICS indicators on the ALS MEDM screen suggested ADC saturations were going on. I had never really looked at the fast time series of the inputs to the phase tracker servos, but these showed saturating behavior on ndscope traces. I went to the LSC rack and measured these on a scope, indeed, they were ~20V pp.

The output of the BeatMouth PDs are going to a ZHL-3A amplifier - we should consider replacing these with lower gain amplifiers, e.g. the Teledyne AP1053. This is relegated to a daytime task.

Other findings tonight:

While working on the PSL table, I somehow put the IMC FSS into a bad state, reminiscent of this behavior. Seems like this is linked to some flaky connection on the PSL table. One candidate is the unstable attachment of the Pomona box between the NPRO PZT and the FSS output - we should install a short BNC cable between these to avoid the lever arm situation we have right now.

[steve, gautam]

We installed some rails to mount the 2U chassis containing ~100m of delay line cabling, and the 1U chassis containing the FET demodulators for the ALS signals in the LSC rack. This has made it MUCH easier for a single person to work there and remove/reinstall these chassis. The delay line box has 100m of cable inside it, and so was rather heavy (~8kg) - previously, it was being supported only by a pair of brackets on the front, so the new arrangement is much more robust. Steve is looking into acquiring plastic spacers of the appropriate width, so that we can secure the units to the rack using usual rack mount screws (but the material of the newly installed rails and the screw heads holding them in place necessitate this plastic spacer). 

Delay line box has been re-installed, demodulator chassis has been removed by me for characterization. Steve will put up photos once the units are re-installed.

For this work, I had to disconnect a bunch of cabling, but only those connected to ALS. All cables were labelled, and I will re-connect them once I am done with the demod chassis.

 I modified the ALS down script. When  the value of  C1:ALS-X(Y)ARM_OUT  becomes larger than the threshold, it turn off the output ON/OFF switch immediately. That is because the ALS servo has ramp time. When script changes the gain to 0, it takes some seconds. That is not good for suspensions.

 After changing servo gain to 0 and turning off the filters, the script waits ramp time and turn on the servo output switch.

I wrote the down script for ALS. This script is  (script)/ALS/ALSdown.py When this script is running, it watches the feedback signal of the ALS control loop so as to shut down the servo immediately when  the suspension is kicked. 

When  the value of  C1:ALS-X(Y)ARM_OUT  becomes larger than the threshold (right now it is 4500 counts), it changes the servo gain to 0, turns off all filters except for FM5 (the filter for phase compensation), resets the history of the phase tracker of each arm and prints the time on window when the suspension kicked

I put the switch on the C1ALS screen, and if you push this switch the window will open (like when you turn on the c1ass script) and the script start to run. For stopping this script, you have to close that window or press Ctrl + C on that window. This is little bit inconvenient, but we will make autolocker script for ALS and this downscript will be included that script soon. So I think it is enough to protect the suspensions right now.

Summary:

Last year, I worked on the ALS delay line electronics, thinking that we were in danger of saturation. The analysis was incorrect. I find that for RF signal levels between -10 dBm and +15 dBm, assuming 3dB insertion loss due to components and 5 dB conversion loss in the mixer, there is no danger of saturation in the I/F part of the circuit.

Details:

The key is that the MOSFET mixer used in the demodulation circuit drives an I/F current and not voltage. The I-to-V conversion is done by a transimpedance amplifier and not a voltage amplifier. The confusion arose from interpreting the gain of the first stage of the I/F amplifier as 1 kohm/10 ohm = 100. The real figures of merit we have to look at are the current through, and voltage across, the transimpedance resistor.  So I think we should revert to the old setup. This analysis is consistent with an actual test I did on the board, details of which may be found here.

We may still benefit from some whitening of the signal before digitization between 10-100 Hz, need to check what is an appropriate place in the signal chain to put in some whitening, there are some constraints to the circuit topology because of the MOSFET mixer.

One part of the circuit topology I'm still confused by is the choice of impedance-matching transformer at the RF-input of this demod board - why is a 75 ohm part used instead of a 50 ohm part? Isn't this going to actually result in an impedance mismatch given our RG405 cabling?

Update: Having pulled out the board, it looks like the input transformer is an ADT-1-1, and NOT an ADT1-1WT as labelled on the schematic. The former is indeed a 50ohm part. So it makes sense to me now.

Since we have the NF1611 fiber coupled PDs, I'm going to try reviving the X arm ALS to check out what the noise is after bypassing the suspect Menlo PDs we were using thus far. My re-analysis can be found in the attached zip of my ipynb (in PDF form).

Summary:

The restoration of the delay-line electronics is complete. The chassis has not been re-installed yet, I will put it back in tomorrow. I think the calculations and measurements are in good agreement.

Details:

Apart from restoring the transimpedance of the I/F amplifier, I also had to replace the two differential-sending AD8672s in the RF Log detector circuit for both LO and RF paths in the ALS-X board. I performed the same tests as I did the last time on the electronics bench, results will be uploaded to the DCC page for the 40m version of the board. I think the board is performing as advertised, although there is some variation in the noise of the two pairs of I/Q readouts. Sticking with the notation of the HP Application Note for delay line frequency discriminators, here are some numebrs for our delay line system:

•   - measured by driving the LO/RF inputs with Fluke/Marconi at 7dBm/0dBm (which are the expected signal levels accounting for losses between the BeatMouth and the demodulator) and looking at the Vpp of the resulting I/F beat signal on a scope. This is assuming we use the differential output of the demodulator (divide by 2 if we use the single-ended output instead).
•  [see measurement]
•  (to be confirmed by measurement by driving a known FM signal with the Marconi)
• Assuming 1mW of light on our beat PDs and perfect contrast, the phase noise due to shot noise is which is ~ 5 orders of magnitude lower than the electronics noise in equivalent frequency noise at 100 Hz.
• The noise due to the FET mixer seems quite complicated to calculate - but as a lower bound, the Johnson current noise due to the 182 ohms at each RF input is ~ 10 pA/rtHz. With a transimpedance gain of 1 kohm, this corresponds to ~10 nV/rtHz. 

In conclusion: the ALS noise is very likely limited by ADC noise (~1 Hz/rtHz frequency noise for 5uV/rtHz ADC noise). We need some whitening. Why whiten the demodulated signal instead of directly incorporating the whitening into the I/F amplifier input stage? Because I couldn't find a design that satisfies all the following criteria (this was why my previous design was flawed):

1. The commutating part of the FET mixer must be close to ground potential always.
2. The loading of the FET mixer is mostly capacitive.
3. The DC gain of the I/F amplifier is low, with 20-30dB gain at 100 Hz, and then rolled off again at high frequencies for stability and sum-frequency rejection. In fact, it's not even obvious to me that we want a low DC gain - the quantity  is directly proportional to the DC transimpedance gain, and we want that to be large for more sensitive frequency discriminating.

So Rich suggested separating the transimpedance and whitening operations. The output noise of the differential outputs of the demodulator unit is <100 nV/rtHz at 100 Hz, so we should be able to saturate that noise level with a whitening unit whose input referred noise level is < 100 nV/rtHz. I'm going to see if there are any aLIGO whitening board spares - the existing whitening boards are not a good candidate I think because of the large DC signal level.

This Hanford alog may be of relevance as we are using the aLIGO AA chassis for the IR ALS channels. We aren't expecting any large amplitude high frequency signals for this application, but putting this here in case it's useful someday.

This test was done, and I determine the frequency discriminant to be  (for an RF signal level of ~2 dBm). 

Attachment #1: Measured and predicted value of the DFD discriminant for a few RF signal levels.

• Methodology was to drive an FM (deviation = 25 Hz, fMod = 221 Hz, fCarrier ~ 40 MHz) with the Marconi, and look at the IF spectrum peak height on a SR785
• The "Design" curve is calculated using the circuit parameters, assuming 4dB conversion loss in the mixer itself, and 3dB insertion loss due to various impedance matching transformers and couplers in the RF signal chain. I fudged the insertion/convertion loss numbers to get this curve to line up with the measurements (by eye).
• For the measurement, I assume the value for FM deviation displayed on the Marconi is an RMS value (this is the best I can gather from the manual). I'll double checking by looking at the RFmon spectrum directly on the Agilent NA.
• X axis calibrated by reading off from the RF power monitor using a DMM and using the calibration data from the bench.
• I could never get the ratio of peak heights in Ichan/Qchan (or the other way around) to better than ~ 1/8 (by moving the carrier frequency around). Not sure I can explain that - small non-orthogonality between I and Q channels cannot explain this level of leakage.

Attachment #2: Measured noise spectrum in the 1Y2 (LSC) electronics rack, calibrated to Hz/rtHz using the discriminant from Attachment #1.

• Something funky with the I channel for X, I'll re-take that spectrum.

I'm still waiting on some parts for the new BeatMouth before giving the whole system a whirl. In the meantime, I'll work on the EX and EY green setups, to try and improve the mode-matching and better characterize the expected suppressed frequency noise of the end NPROs - the goal here is to rule out the excess low-frequency noise that was seen in the ALS signals coming from unsuppressed frequency noise.

Bottom lines: 

1. The DFD noise is at the level of ~ 10mHz/rtHz above 10 Hz. This justifies the need for whitening before ADC-ing.
2. The measured signal/noise levels in the DFD chain are in good agreement with the "expected" levels from circuit component values and typical insertion/conversion loss values.
3. Why are there so many 60 Hz harmonics???

Attachment #1 shows the schematic of the test setup. Signal generator (Marconi) was used to supply the RF input. We observed the IF output in the following three test conditions.

1. Observed the spectrum with FM modulation (fcarrier of 40 MHz and fmod of 221 Hz )- a peak at 221 Hz was observed.
2. Observed the noise spectrum without FM modulation.
3. Observed the noise spectrum after disconnecting the delayed output of the delay line. 

• It is observed that the broad band noise level is higher without FM modulation (2) compared to that we observed after disconnecting the delayed output of the delay line (3).
• It is also observed that the noise level is increasing with increase in RF input power. 
• We need to find the reason for increase in broad band noise .

[Paco, Anchal]

One of the crucial and currently limiting calibration errors is in the ALS beat. We think a major driver of calibration uncertainty may be the DFD TF calibration since it depends on the RF beat power.

The RF beat power as monitored by C1:ALS-BEATY_RF_POW shows relative fluctuations of 0.02% at the 16 Hz sampling rate (actually 0.09% rms from Attachment #1) once the single arm and YAUX laser are both locked. This sounds ok, but that's just a statistical estimate. The beat frequency changes every time we break and reacquire the lock, making the BEAT PD frequency response and DFD overall calibration change their nominal TF. This changes can be significantly larger than 0.02% (e.g 4.9% = observed change in the RF power after breaking and reacquiring the YAUX lock this morning). This is far more significant than the contribution from the RIN on the two lasers.

This kind of systematic error could be addressed by either/all:

• Stabilizing the ALS BEAT absolute frequency (offset locking using freq counter and simple integrator or equivalent)
• Rejecting RF beat amplitude fluctuations by mixing a stable DC voltage in the DFD, or using a comparator. 
  • Our plan this afternoon is to quickly measure the RF AM rejection level from a simple mixer and a stable voltage reference and plan ahead.
• ISS on both lasers

Eventually the instability in the Y end PDH servo turned out to be some kind of an alignment issue.

After carefully realigning the green beam to the Y arm, the UGF of the ALS loop became able to be at more than 50 Hz.

With this UGF it became able to suppress the arm motion to the ADC noise level (few 100 pm in rms).

Now I am scanning the arm length to look for a TEM00 resonance.

(the Story)

I have noticed that the spatial fringe pattern of the reflected green light was very sensitive to the pitch motion of ETMY when the green light was locked to the Y arm.

So I realigned the last two launching mirrors to minimize the reflected light. Indeed the misalignment was mainly in the pitch direction.

I basically translated the beam upward by a couple of mm or so.

The amount of the DC reflection is about 2.4 V when it is unlocked and it is now 0.77 mV when the green light is locked.

[Paco]

I took ~ 7 minutes of XALS beatnote data with the XAUX laser locked to the XARM cavity, and the XARM locked to PSL to develop an allan deviation estimator. The resulting timeseries for the channel C1:ALS-BEATX_FINE_PHASE_OUT_HZ_DQ (decimated timeseries in Attachment #1) was turned into an allan variance using the "overlapped variable tau estimator":

Where  represents the k-th data point in the raw timeseries, and are the variable integration intervals under which two point variances are computed (the allan variance is a special case of M-point variance, where M=2). Then, the allan deviation is just the square root of that. Attachment #2 shows the fractional deviation (normalized by the mean beat frequency ~ 3 MHz for this measurement) for 100 integration times spanning the full duration (~ 7 min = 420 s).

The code used for this lives in Git/40m/labutils/measuremens/ALS/

If this estimate is any good, wherever the fractional beatnote deviation reaches a minimum value can be used as a proxy for the longest averaging time that give a statistical increase in SNR. After this timescale, the frequency comparison is usually taken over by "environmental instabilities" which I don't think I can comment further on. In our particular estimate, the 100 second integration gives a fractional deviation of ~ 0.44 %, or absolute deviation of 12.925 kHz.

what's the reasoning behind using df/f_beat instead of df/f_laser ?

I guess it didn't make sense since f_beat can be arbitrarily moved, but the beat is taken around the PSL freq ~ 281.73 THz. Attachment #1 shows the overlapping tau allan deviation for the exact same dataset but using the python package allantools, where this time I used the PSL freq as the base frequency. This time, I can see the minimum fractional deviation of 1.33e-13 happening at ~ 20 seconds.

Another, more familiar interpretation

The allan variance is related to the beatnote spectral density as a mean-square integral (the deviation is then like the rms) with a sinc window.

[Masayuki, Manasa] 

Flowchart for ALS autolocker. The error signal thresholds will be decided by trial and error.

  I think we can use the IMC autolocker to start with getting this started. Once Jamie fixes the NDSSERVER environment variable bug, we should be able to use his more slick automation code to make it auto lock.

Step by step procedure for stabilizing arms using ALS servo:

The procedure is the same for both the arms. 

0. Check that the ALS arm servos are turned OFF and not sending any signals to the ETM suspensions. 

1. Find the beat note by varying the laser temperature (moving the slider for SLOW_SERVO2_OFFSET).
Tip: It is easier to have the arms locked using IR PDH while searching for the beat note. Also check the stability of the MC. Unstable MC will cause the PSL temperature to drift and thereby affect the beat frequency.

2. Once you have the beat note, check if the beat amplitude is ~ -15 to -20 dBm. If the amplitude is small, then the alignment needs to be fixed (either the green input pointing at the end tables or the PSL green alignment). This is important because the UGF of the phase tracking loop (should be above 1KHz) changes with the amplitude of the beat note.
Also the beat frequency should be < 100 MHz; preferably below 80 MHz.

3. Disable IR PDH locking if you had used it while searching for the beat note. 

4. Press CLEAR HISTORY button for the phase tracker servo. Check if the phase tracking loop is stable (phase tracker servo output counts should not be ramping up). If the phase tracker is not stable, check the servo gain and phase margin of the loop.

5. Turn OFF all filters in the ALS arm servo filter module except for FM5 (phase compensation filter). With ALS arm servo gain set to zero, enable the arm servo and allow ALS control signals to be sent to the ETM suspensions.

5. Open dtt and look at the power spectrum of the ALS error signal (C1:ALS-BEAT?_FINE_PHASE_OUT_HZ). 

6. Set ALS arm servo gain +/- 0.1 to check the sign of the servo gain. Wrong sign of gain will make the loop unstable (beat note moving all over the frequency range on the spectrum analyzer). If this happens, set the gain to zero immediately and clear history of phase tracker servo. If you have set the correct sign for gain, the servo will stabilize the beat note frequency right away. 

7. Once you know the correct sign of the servo gain, increase the gain in steps while simultaneously looking at the power spectrum of error signal on dtt (it is convenient to set dtt measurements to low bandwidth and exponential measurement settings). Increase the gain until you can see a slight bump close to the UGF of the ALS servo (>100Hz). 
There have been times when this servo gain was in a few hundreds; but right now it varies from +/- 10-20 for both the arms. So we are stepping up gain in steps of +/- 2.

8. Enable filters (FM2, FM3, FM6, FM7, FM8). Wait to see the rms noise of the error signal go down (a few seconds).

9. Enable boost filter (FM10). There also exists a weaker boost filter (FM4) which we don't use any more. 

Note:

1. Beat frequency changes affect both the servo gain and sign of gain. So if you lose stability of ALS servo at any point, you should go through all the steps again.

2. At any point if the ALS arm servo becomes unstable (which can happen if the MC loses lock or if the beat frequency was too high ), change the servo gain to zero immediately. Turn OFF all the filters except for FM5 (if they were enabled) and reset phase tracker servo (CLEAR HISTORY button in the phase tracker filter module). Masayuki has written the down script that does all this. The script will detect arm servo loop instability by continuously looking at the feedback signal. Details about the script can be found here. 

Here is a cheat sheet that can give you an idea of the SLOW SERVO2 offset range to scan in order to find the beat note:

PSL temperature  X offset   Y offset
31.58                     5278       -10890
31.52                     5140       (not recorded)
31.45                     4810       (not recorded)
31.33                     4640       -10440
31.28                     4500       -10340
            

After aligning the arms to IR, I aligned the Y green beam to the arm.  Also, the X green beatnote was very small, so I aligned the PSL green for X.

[Koji Annalisa]

We did the same mod of the beatbox for the Y arm too. See
http://nodus.ligo.caltech.edu:8080/40m/8855

[Manasa Koji]

Summary:
The new whitening filters improved the out-of-loop ALS stability of the Y arm down to 300Hz (20pm_rms in displacement).

- After modifying the whitening filters, the out-of-loop stability of the arms were tested with the IR PDH signals.

- The X arm showed non-stationarity and it made the ALS servo frequenctly fell out of lock.

- For now we decided to use the Y arm for the PRMI+one arm trial.

- The performance of the ALS was tested with several measurements. (attachment 1)

Cyan: Stability of the beatnote frequency with the MC and the arm freely running. The RMS of the day was ~6MHz.

Blue: Sensing limit of the beat box was tested by giving a signal from Marconi. The same amplitude as the X arm beat was given as the test signal.
This yielded the DC output of ~1200 counts.

Green: Out-of-loop estimation of the beatbox performance. This beat note stability was measured by controlling the arm with the IR PDH signal.
Assuming the PDH signal has better SNR than the beat signal, this gives us the out-of-loop estimation of the stability below 150Hz, which is the
unity gain frequency of the ALS loop.
Above 150Hz the loop does not force this noise to the suspension. Just the noise is injected via a residual control gain (<1).

Black: In-loop evaluation of the ALS loop. This becomes the left over noise for the true stability of the arm (for the IR beam).

Red: The arm was brought to the IR resonance using the ALS offset. The out-of-loop stability was evaluated by the IR PDH signal.
This indeed agreed with the evaluation with the other out-of-loop evaluation above (Green) below 150Hz.

Attachment 2 shows the time series data to show how the arm is brought to the resonance.
1 count of the offset corresponds to ~20kHz. So the arm started from 200kHz away from the resonance
and brought to the middle of the resonance.

(Manasa downloaded the 2k sampled data so that we can use this for presentations.)

Awesome !

 Path to data (retreived using getdata)

/users/manasa/data/130717/YALS_scan

I touched up the alignment of the Ygreen on the PSL table.

Please check the X&Y ALS out-of-loop stability. Use fine resolution (BW0.01). Calibrate the POX/POY in Hz.

I have hooked the ALS beatbox into the c1ioo DAQ.  In the process, I did some rewiring so that the channel mapping corresponds to what is in the c1gcv model.

The Y-arm beat PD is going through the old proto-DFD setup.  The non-existant X-arm beat PD will use the beatbox alpha.

Y coarse I (proto-DFD) --> c1ioo ADC1 14 --> C1:ALS_BEATY_COARSE_I
Y fine   I (proto-DFD) --> c1ioo ADC1 15 --> C1:ALS_BEATY_FINE_I
X coarse I (bbox alpha)--> c1ioo ADC1 02 --> C1:ALS_BEATX_COARSE_I
X fine   I (bbox alpha)--> c1ioo ADC1 03 --> C1:ALS_BEATX_FINE_I

This remapping required coping some filters into the BEATY_{COARSE,FINE} filter bank.  I think I got it all copied over correctly, but I might have messed something up.  BE AWARE.

We still need to run a proper cable from the X-arm beat PD to the beatbox.

I still need to do a full noise/response characterization of the beatbox (hopefully this weekend).

I. The Y arm stayed stable through last night and I have saved the arm lock settings to IFOconfigure.

II. ALS X arm noise measurements

I looked at the before and after noise of ALSX.

Settings:
Phase tracker gain = 85
Xarm servo gain = -17

The rms in loop noise has dropped from 3KHz to 500 Hz.

Attachment 1: Phase tracker OLTF

Attachment 2: Free running noise and in loop noise

Attachment 3: Out of loop noise (measured with arms locked using PDH for IR)

Attachment 4: ALS arm servo OLTF

xml data files can be found in /users/manasa/data/140430/

[Rana, Jenne]

We have looked at a few things that do and don't affect the out of loop noise of the ALS X beat, and found that cavity alignment and beatnote RF frequency had the strongest effects.

Possible causes of noise:

1.  Air currents from A/C or flowbench.  No effect. 

        * When table lid is on, turning on and off the flow bench air did not qualitatively change the out of loop beatnote time series signal.

2.  Scattered light from other beams hitting green PDH PD.  No effect.

        * There are a few spots of green light that are hitting the case of the PDH photodiode, but when I put an iris in place to block those spots, there was no change in the beatnote spectra.  This makes sense to me since none of those spots were close to hitting the diode itself. 

        * Rana did notice that the beam was not well centered on the PD, so he steered the beam onto the center of the diode.  Also, the PD is now tilted a little bit so that the reflection from the diode doesn't go back into the beam path.  Neither of these things had an effect that we noticed in the beatnote noise.

3.  Oplev laser light getting to PDH PD.  Not tested.

       * We don't see any red light over by the PDH PD, so we did not try turning off the oplev's laser to see if that had an effect, but we suspect that it is not the cause of our noise.

4.  Clipping of main IR / green beam on Xend table.  Not tested.

      * We should still go have a look at this, but we no longer think that this is the main cause of the elevated noise.

5.  Scattered light all over Xend table.  Not tested.

     * We should still work on dumping extraneous beams on the table, but we do not think that this is the main cause of the elevated noise.

     * Rana took some photos so that we can see how truely bad the situation is.

6.  Amplitude modulation dip in NPRO.  Not tested.

    * It is probably still a good idea to check this, in case the dip in the amplitude modulation has changed over the year or two since it was last measured, but we also don't think that this is the main problem.

7.  Check PDH servo.  Not done.

     * I think this is still on Q's long-term todo list, but we should give the PDH servos a once-over.

8.  Arm cavity longitudinal motion.  No effect.

     * While the Xarm was locked with IR, we put a line at 1.7 Hz with 325 counts into the ITMX position.  To keep lock, the ETM had to move as well.  When we turned on this line (and increased its amplitude up to the final value of 325 cts), we did not see any qualitative change in the beatnote time series noise.

9.  Arm cavity alignment.  Significant DC effect.

    * When the alignment of one of the arm cavity mirrors is changed, the DC value of the beatnote signal changes. 

           * ITMX moved in yaw, we see a 7kHz/15urad DC shift in the BEATX_FINE_PHASE_OUT_HZ time series.

          * ETMX moved in yaw, we see an 8kHz/5.5urad DC shift in the time series.  We aren't sure why this is about a factor of 3 times larger effect (same shift for smaller misalignment) than the ITM.

    * We want to do a Yuta-style analysis to see what the angle to length coupling looks like, so that we can measure the angular motion of our cavity mirrors and put the expected noise into our ALS noise budget.  Perhaps this will help us understand the low frequency difference between our in-loop beatnote error signal and our in-loop PDH error signals (red vs. maroon on the ALS noise budget posted above Pianosa). 

    * I've asked Manasa to take some transfer functions in the morning, so that we can start to have an idea of what is going on with this.

10.  Beatnote RF frequency.  Significant broadband effect.

     * We have found that when the Xarm beatnote is at low RF frequencies, the noise is high, and when the beatnote is at high RF frequencies, the noise is low! 

     * Low RF freqs are below about 40 MHz, while high RF freqs are above about 90 MHz.  This has not been tested for the Yarm.  Also, these are for the case of "temp slider up, beatnote up".  I have not checked if the same is true for the other side of the PSL frequency, although I don't have reason to believe that it would be.

     * Maybe we are saturating some amplifiers?  We need to check this out.  One thought that Den mentioned was the harmonics, and that perhaps they are causing trouble in the electronics.

     * Den is going to think about implementing a frequency divider so that we can directly digitize the beatnote signal. 

    * Here are spectra for different cases:

        * And here is a spectrogram showing us going back and forth between the high and low noise states:

                     *  A:  First noticing that noise is good when RF frequency is high.

                     * B:  Not locked on TEM00 mode, so extra noisy.  Disregard.

                     * C:  Bad noise time.  Xbeat was 21 MHz (dark purple on DTT spectrum above), Ybeat was 118 MHz (sea green on DTT spectrum above).

                    * D:  Good noise time. Xbeat was 89 MHz (light purple on DTT plot), Ybeat was still 118MHz (turquoise on DTT plot).

                     * E:  Bad noise time.  Xbeat was 37.5 MHz, Ybeat was still 118 MHz.

                     * F:  Good noise time.  Xbeat was 113 MHz, Ybeat was still 118 MHz.

Attached is the measurement of the transfer function from ITMX oplev error in yaw to the ALSX error signal.

The arm was locked to the IR using POX and the green beat frequency (between X arm trans in green and PSL green) in this case was 27MHz.

The transfer function looks mostly flat between 1Hz - 30Hz at 700-800 Hz/urad. The DC shift that Jenne measured from the time series is ~500 Hz/urad.

So far I have not been able to measure the TF below 1Hz without the arm losing its lock. Updates will follow.

Data xml file can be found in /users/manasa/data/140521/

Below are the transfer functions measured between the angular (pit, yaw) motion of X arm mirrors and the ALSX error signal. The measurements were again made for 1Hz-30Hz.

The transfer functions are mostly flat.

ITMX P - 200-300 Hz/urad (beat freq = 45 MHz)

ITMX Y - 700-800 Hz/urad (beat freq = 27MHz)

ETMX P - 500-600 Hz/urad (beat freq = 56 MHz)

ETMX Y - 1000-2000 Hz/urad (beat freq = 62.5MHz)

Data xml files can be found in /users/manasa/data/140521/

I saw big misalignment on the GTRX camera, I went to the PSL table and aligned the beat beams.

I disconnected the RF out of the X beat PD and  connect an oscilloscope.
The beat amplitude was 15mVpp at the beginning and is 60mVpp right now.
I checked the alignment on this RF PD and the DC PD as well as the spot on the CCD.

The RF cable was connected again.

Jenne and I ran the ALS and scanned the arm cavity. We had the impression that the noise level of the ALS improved,
but I don't have correctly calibrated measurement. Let's do it tomorrow in the day time.

The Yarm beat alignment look awful. We should align this too.

The X and Y arms were locked successfully using ALS and the arms could be scanned and held to support IR resonance.

The same procedure as in elog 9219 was followed. In-loop noise was measured to be between 200-300 Hz rms for the lock.

ALS settings for the lock

X arm : FM 2, 3, 5, 6, 7, 8, 10  Gain = 11.0
Y arm : FM 2, 3, 5, 6, 7, 8, 10  Gain = 10.0

Nice restoration. We eventually want to make transition of the servo part from ALS to LSC model for the further handing off to the other signals.
Please proceed to it.

Q and I have started to use the ALS slow servo for the end aux lasers while locking the arms using ALS. The servo prevents us from hitting the limits of the PZT range for the end lasers and a better PDH locking.

But keeping the servo ON causes the slow output to drift away making it hard to find the beat note everytime the arm loses lock. The extensive beat note search following the unlock can be avoided by clearing history of the slow servo.

 Ha!

A few more related changes:

1. The couplers that used to sit on the green beat PDs on the PSL table have now been shifted to the IR broadband PDs in the FOL box so that I can get the IR beat frequency over to the frequency counters. The FOL box itself, along with the fibers that bring IR light to the PSL table, have been relocated to the corner of the PSL table where the green beat PDs sit because of cable length constraints.
2. I've updated the ALS slow control MEDM screen to allow for slow control of the beat frequency. The servo shape for now is essentially just an integrator with a zero at 1 Hz. The idea is to set an offset in the new filter module, which is the desired beat frequency, and let the integrator maintain this beat frequency. One thing I've not taken care of yet is automatically turning this loop off when the IMC loses lock. Screenshot of the modified MEDM screen is attached. 
3. I checked the performance by using the temperature sliders to introduce an offset. The integrator is able to bring the beat frequency back to the setpoint in a few seconds, provided the step I introduced was not two big (~20 counts, but this is a pretty large shift in beat frequency, nearly 20MHz).

To do:

1. Figure out how to deal with the IMC losing lock. I guess this is important if we want to use the IR beatnote as a diagnostic for the state of the X AUX laser.
2. Optimize the servo gains a little - I still see some ringing when I introduce an offset, this could be avoided...

Here is our first attempt at a single-arm noise budget for ALS.

Attachment 1 shows the loop diagram we used to calculate the contribution of different noises.

Attachment 2 shows the measured noise at C1:ALS-BEATX_PHASE_FINE_OUT_HZ when XARM was locked to the main laser and Xend Green laser was locked to XARM.

• The brown curve shows the measured noise.
• The black curve shows total estimated noise from various noise sources (some of these sources have not been plotted as their contribution falls off the plotting y-lims.)
• The residual frequency noise of Xend green laser (AUX) is measured by measuring the PDH error monitor spectrum from C1:ALS-X_ERR_MON_OUT_DQ. This measurement was converted into units of V by multiplying it by 6.285e-4 V/cts. This factor was measured by sending a 43 Hz 100 mV sine wave at the readout point and measuring the output in the channel.
• This error signal is referred to AUX_Freq input in the loop diagram (see attachment 1) and then injected from there.
• All measurements were taken to Res_Disp port in the 'Out-of-Loop Beat Note' block (see attachment 1).
• In this measurement, we did not DAC noise that gets added when ALS loop is closed.
• We added ADC noise from Kiwamu's ALS paper after referring it to DFD input. DFD noise is also taken from Kiwamu's ALS paper data.

Inference:

• Something is wrong above 200 Hz for the inclusion of AUX residual displacement noise. It is coming off as higher than the direct measured residual noise, so something is wrong with our loop diagram. But I'm not sure what.
• There is a lot of unaccounted noise everywhere from 1 Hz to 200 Hz.
• Rana said noise budget below 1 Hz is level 9 stuff while we are at level 2, so I'll just assume the excess noise below 1 Hz is level 9 stuff.
• We did include seismic noise taken from 40m noise budget in 40m/pygwinc. But it seems to affect below the plotted ylims. I'm not sure if that is correct either.

Unrelated questions:

• There is a slow servo feeding back to Green Laser's crystal temperature by integrating PZT out signal. This is OFF right now. Should we keep it on?
• The green laser lock is very unreliable and it unlocks soon after any signal is being fed back to the ETMX position.
• This means, keeping both IR and green light locked in XARM is hard and simultaneous oscillation does not last longer than 10s of seconds. Why is it like this?
• We notice that multiple higher-order modes from the green laser reach the arm cavity. The HOMs are powerful enough that PDH locks to them as well and we toggle the shutter to come to TEM00 mode. These HOMs must be degrading the PDH error signal. Should we consider installing PMCs at end tables too?

Here's an updated X ARM ALS noise budget.

Things to remember:

• Major mistake we were making earlier was that we were missing the step of clicking  'Set Phase UGF' before taking the measurement.
• Click the clear phase history just before taking measure.
• Make sure the IR beatnotes are less than 50 MHz (or the left half of HP8591E on water crate). The DFD is designed for this much beatnote frequency (from Gautum).
• We took this measurement with old IMC settings.
• We have saved a template file in users/Templates/ALS/ALS_outOfLoop_Ref_DQ.xml . This si same as ALS_outOfLoop_Ref.xml except we changed all channels to _DQ.

Conclusions:

• Attachment 1 shows the updated noisebudget. The estimated and measured RMS noise are very close to eachother.
• However, there is significant excess noise between 4 Hz and 200 Hz. We're still thinking on what could be the source of these.
• From 200 Hz to about 3 kHz, the beatnote noise is dominated by AUX residual frequency noise. This can be verified with page 2 of Attachment 2 where coherence between AUX PDH Error signal and BEATX signal is high.
• One mystery is how the measured beatnote noise is below the residual green laser noise above 3 kHz. Could this be just because the phase tracker can't measure noise above 3kHz?
• We have used estimated open loop transfer function for AUX from poles/zeros for uPDH box used (this was done months ago by me when I was working on ALS noise budget from home). We should verify it with a fresh OLTF measurement of AUX PDH loop. That's next on our list.

[Meenakshi, Gautam, Shruti]

Summary:

- We initially aligned the arm cavities to get the green lasers locked to them. For the X arm cavity, we tweaked the ITMX and ETMX pitch and yaw and toggled the X green shutter until we saw something like a TEM00 mode on the monitor and a reasonable transmitted power.

- With the LSC servo enabled, the IR light also became resonant with the cavities.

- Then we measured the noise in different configurations. Attachment 1 shows the the ALS OOL (in the IR beat signal) noise with the arms locked inidividually via PDH.

The script for plotting the ALS beat frequency noise is:

[Meenakshi, Shruti]

Even though we were not able to lock the the IR beat (by enabling LSC) during the day possibly because of increased seismic activity, we tried to the measure the ALS beat frequency noise by changing the PDH side-band frequency to different values.

I tried choosing values that corresponded to the peaks in the PM/AM as found in elog:15206 but for some reason unknown to us the cavity did not lock between 700-800 kHz.

The three attachments have data for different sideband frequencies:

Attachment 1: 819.472 kHz (peak in PM/AM, measured around noon)

Attachment 2: 225.642 kHz (peak in PM/AM, measured earlier in the morning)

Attachment 3: 693.500 kHz (not a peak in PM/AM)

We don't think these plots mean much and will do the measurement at some quieter time more systematically.

While doing the experiment, the ITMY pitch actuation was changed from -2.302 to -2.3172V because it locked better.

The ITMX, ETMX alignment was also tweaked to try to lock with different sideband frequencies and then restored to the values that were found earlier in the morning.

Unlikely, the alignment was probably just not good. I restored the alignment and now the arms can be locked to IR frequency.

We proceeded with the trying to measure the ALS out-of-loop noise of the X arm when the X arm cavity is green-locked using different PDH sideband frequencies.

Before doing the experiment, Koji helped us with getting the arm cavities locked in IR using LSC (length) and ASC (angular).

With the arms locked in IR and green, we repeated the same measurements as before at different sideband frequencies (Refer Attachment 1 - label in Hz). We did not optimize the phase nor did we look at the PDH error signal today which is possibvly why we did not see an improvement in the noise. We will look into this possibly tomorrow.

Could you please put physical units on the Y-axis and also put labels in the legend which give a physical description of what each trace is?

It would also be good to a separate plot which has the IR locking signal and the green locking signal along with this out of loop noise, all in the same units so that w can see what the ratio is.