1.

> Looking through the manual, I found a recommendation (pg10) that the "IN-" terminal of the Acromag ADC units be tied to the "RTN" pins on the same units. I don't know if this preserves the differential receiving capability of the Acromag ADCs

I suppose, we loose the differential capability of an input if the -IN is connected to whatever defined potential. We should check if the channels are still working as a true differential or not.

2. If the multi bit operation is too complicated to solve, we can use EPICS Calc channels to breakout a value to bits and send the individual bits as same as the other individual binary channels.

As with the last two posts, I eliminated more unnecessary passing through c1rfm for IPC connections between c1mcs and c1ioo.

All models were rebuilt/installed/restarted and svn committed.  Everything is working and we have eliminated almost all IPC errors and significantly simplified things.

Jordan and I removed another 10 kg of cabling from 1X2. The c1iool0 crate now has all cabling to it disconnected - but it remains in the rack because I can't think of a good way to remove it without disturbing a bunch of cabling to the fast c1iool0 machine. We can remove it the next time the vertex FEs crash. Cross connects have NOT been removed - we will identify which cross connects are not connected to the fast system and trash those. 

Do we want to preserve the ability to use the PZT driver in 1X2?

After discussing with Koji, I removed the PZT driver and associated AI card from the Eurocrate at 1X2. The corresponding backplane connectors were also removed from the cross connects. An additional cable going from the DAC to IDC adaptor on 1X2 was removed. Finally, some cables going to the backplane P1 and P2 connectors for slots in which there were no cards were removed. 

Finally, there is the IMC WFS whitening boards. These were reconfigured in ~2016  by Koji to have (i) forever whitening, and (ii) fixed gain. So the signals from the P1 connector no longer have any influence on the operation of this board. So I removed these backplane cables as well.

Some pics attached. The only cross connect cabling remaining on the south side of 1X2 is going to the fast BIO adaptor box - I suspect these are the triggered fast whitening switching for the aforementioned WFS whitening board. If so, we could potentially remove those as well, and remove all the cross connects from 1X1 and 1X2.

Update 1720: indeed, as Attachment #2 shows, the RTCDS BIO channels were for the WFS whitening switching so I removed those cables as well. This means all the xconnects can be removed. Also, the DAC and BIO cards in c1ioo are unused.

We are going to replace the old Sun c1ioo with a modernized supermicro. At the opportunity, remove the DAC and BIO cards to use them with the new machines. BTW I also have ~4 32ch BIO cards in my office.

Summary:

1. Swapping out the KEPCO HV supplies (linear) I was using for a pair of HP6209s I borrowed from Rich has improved the noise performance somewhat.
2. However, there is still an excess relative to the model. I confirmed that this excess originates from the PA95 part of the circuit (see details).
3. The bypass capacitors don't seem to have any effect on the measured ripple from these HP6209s. Maybe they're internally fitted with some 10uF or similar bypass caps?
4. The production version of this board, with several improvements (after discussions with Koji and Rich), are on the DCC. They're being fabbed right now and will arrive in ~1 week for more bench testing. 

Power supply bypassing [updated 10pm]:

As mentioned earlier in this thread, I prepared a box with two 10uF, 1kV rated capacitors to bypass the high-voltage rails (see inset in the plot), to see if that improves the performance. However, in measuring the voltage ripple directly with the SR785 (no load connected), I don't see any significant difference whether the decoupling caps are connected or not, see Attachment #1. For this, and all other HV measurements made, I used this box to protect the SR785. One hypothesis is that this box itself is somehow introducting the excess noise, maybe because of leakage currents of the diode pair going into the 1Mohm SR785 input impedance, but I can't find any spec for this, and anyway, these diodes should be at ground potential once the transient has settled and the DC blocking capacitor has charged to its final value.

Note that the 10uF caps have an ESR of 7.2 mOhms. The HP6209 has a source impedance "<20mOhm" when operated as a CV source, per the datasheet. So perhaps this isn't so surprising? The same datasheet suggests the source impedance is 500 mOhms from 1kHz to 100 kHz, so we should see some improvement there, but I only measured out to 2 kHz, and I didn't take much effort to reduce these crazy peaks so maybe they are polluting the measurement out there. There must also be some continuous change of impedance, it cannot be <20 mOhm until 1 kHz and then suddenly increase to 500 mOhms. Anyways, for this particular circuit, the nosie DC-1kHz is what is important so I don't see a need to beat this horse more. 

Simplified circuit testing:

I decided to see if I can recover the spec'd voltage noise curve from the PA95 datasheet. For this, I configured the PA95 as a simple G=31 non-inverting amplifier (by not stuffing the 15 uF capacitor in the feedback path). Then, with the input grounded, I measured the output voltage noise on the circuit side of the 25kohm resistor (see inset in Attachment #2). To be consistent, I used the DC blocking box for this measurement as well, even though the output of the PA95 under these test conditions is 0V. Once again, there is considerable excess around ~100 Hz relative to a SPICE model. On the basis of this test, I think it is fair to say that the problem is with the PA95 itself. As far as I can tell, I am doing everything by the book, in terms of having gain > 10, using a sufficiently large compensaiton cap, HV rail decoupling etc etc. Note that the PA95 is a FET input opamp, so the effects of input current noise should be negligible. The datasheet doesn't provide the frequency dependence, but if this is just shot noise of the 1200 pA input bias current (for 300 V rails, per the spec), this is totally negligible, as confirmed by LTspice.

In the spirit of going step-by-step, I then added the feedback capacitor, and still, measured noise in excess of what I would expect from my model + SR785 measurement noise.

Integrated circuit testing:

After the above simplified test, I stuffed a full channel as designed, and tested the noise for various drive currents. To best simulate the operating conditions, an Acromag XT1541 was used to set the DC voltage that determines the drive current through the 25 kohm resistor. The measurements were made on the circuit side of this resistor (I connected a 20ohm resistor to ground to simulate the OSEM). As shown in Attachment #3, the noise with these HP6209 supplies is significantly better than what I saw with the KEPCO supplies, lending further credence to the hypothesis that insufficient PSRR is the root of the problem here. I've added subplots in a few different units - to be honest, I think that reaching this level of measured displacement noise at the 40m at 100 Hz would already be pretty impressive.

So what's next?

The main design change is that a passive R-C-R (4k-3uF-20k) replaces the single 25kohm resistor at the output of the PA95. 

• This allows similar current drive range.
• But adds an LPF to filter out the observerd excess noise at 100 Hz. 

Let's see if this fixes the issue. Not that I've also added a pair of input protection diodes to the input of the PA95 in the new design. The idea is that this would protect the (expensive) PA95 IC from, for example, the unit being powered with the +/- 18V rail but not the +/- 300 V rail. As I type this, however, I wonder if the leakage current noise of these diodes would be a problem. Once again, the datasheet doesn't provide any frequency dependence, but if it's just the shot noise of the 1nA expected when the diodes are not reverse biased (which is the case when the PA95 is operating normally since both inputs are at nearly the same potential), the level is ~20 fA/rtHz, comparable to the input current noise of the PA95, so not expected to be an issue. In the worst case, the PCB layout allows for this component to just be omitted. 

I generated the following plots from the two sets of huddle test data we have for the accelerometers. 

Old data: 6 accelerometers, no cables clamped, no box, 55 mins worth of data.

New data: 3 accelerometers, cables clamped, foam box put on placed and completely sealed, 20 mins worth of data.

I made sure to use the same Impuse response time (6 sec) and sampling frequency (256 Hz), as well as every other parameter for the calculations.

Top left: The resultant self noise curve using the new data, there is definitely and improvement in the 0.5-2 Hz band. 

Top right: Resultant self noise using the old data, for the first set of three accelerometers.

Bottom left: Old data result for the remaining three accelerometers.

Bottom right: Old data result, using all six accelerometers as witnesses instead.

There were many locklosses from the point where the arm powers were somewhat stabilized. Attachments #1 and #2 show two individual locklosses. I think what is happening here is that the BS seismometer X channel is glitching, and creating a transient in the angular feedforward filter that blows the lock. The POP QPD based feedback loop cannot suppress this transient, apparently. For now, I get around this problem by boosting the POP QPD feedback loop a bit, and then turning the feedforward filters off. The fact that the other seismometer channels don't report any transient makes me think the problem is either with the seismometer itself, or the readout electronics. The seismometer masses were recently recentered, so I'm leaning towards the latter.

I didn't explicitly check the data, but I am reasonably certain the same effect is responsible for many PRMI locklosses even with the arms held off resonance (though the tolerance to excursions there is higher). Pity really, the feedforward filters were a big help in the lock acquisition...

To study the evilution of the AO path TFs a bit more, I've hooked up POY11_Q Mon to IN1 of the CM board. I will revert the usual setup later in the evening.

Update 1730: I've returned the cabling at 1Y2 to the nominal config, and also reverted all EPICS settings that I modified for this test. Y-arm POY locking works. Attachment #1 shows the summary of the results of this test - note that the AO gain was kept fixed at +5dB throughout the test. I have arbitrarily trimmed the length of the frequency vector for some of these traces so that the noisy measurement doesn't impede visual interpretation of the plots so much. At first glance, the performance is as advertised. I basically followed the settings I had here to get started, and then ramped up various gains to check if the measured OLTF evolved in the way that I expected it to. The phase lead due to the AO path is clearly visible.

Some important differences between this test and the REFL11 blending is (i) in the latter case, there will also be a parallel loop, CARM_A, which is effecting some control, and (ii) the optical gain of CARM-->REFL11_I is much higher than L_Y-->POY. So the initial gain settings will have to be different. But I hope to get some insight into what the correct settings should be from this test. I think IMC servo IN2 gain and AO gain slider on the CM board are degenerate in the effect they have, modulo subtle effects like saturation.

One possibility is that the gain allocation I used yesterday was wrong for the dynamic range of the CARM error signal. In some initial trials today, when I set the CM board IN1 gain to -32dB (as in the case of attempting the CARM RF handoff) and compensated for the reduced POY PDH fringe amplitude by increasing the digital gain for the CM_Slow path, I found that there was no phase advance visible even when I ramped up the IMC IN2 gain to +10dB. So, for the CARM handoff too, I might have to start with a higher CM board IN_1 gain, compensate by reducing the CM_Slow digital gain even more, and then try upping the IMC IN2 gain.

P.S. When the excitation input to the CM board was enabled in order to make TF measurements, I saw significant increase in the RMS of the error signal. Probably some kind of ground loop issue.

This measurement tells you how the gain balance between the SLOW_CM and AO paths should be. Basically, what you need is to adjust the overall gain before the branch of the paths.

Except for the presence of the additional pole-zero in the optical gain because of the power recycling.

You have compensated this with a filter (z=120Hz, p=5kHz) for the CM path. However, AO path still don't know about it. Does this change the behavior of the cross over?

If the servo is not unconditionally stable when the AO gain is set low, can we just turn on the AO path at the nominal gain? This causes some glitch but if the servo is stable, you have a chance to recover the CARM control before everything explodes, maybe?

Summary of discussion between Koji and gautam on 14 Oct:

1. Koji questioned the accuracy of the "open loop" ASD shown here. While it may not be entirely accurate to compute the free-running (homodyne) phase noise simply by taking the arctangent of the I and Q signals (because the magnitude of the signal is also changing), gautam claims the estimate is probably still close to the true homodyne phase, especially since the ratio of the "in-loop" and free-running ASDs gives something that closely approximates the magnitude of the supposed OLG of the system.
2. Koji suggested the following tests:
   • Investigate the relative stability of the two RF signal generators involved in this system. Since the 44 MHz electrical LO signal (for demodulation) is generated by a separate IFR from the one used to imprint 11 MHz and 55 MHz phase modulation sidebands on the main PSL beam, we want to investigate what the drift is.
   • Try implementing an analog feedback loop using LB1005 - the idea being we should be able to implement higher bandwidth control, for better suppression of the high frequency noise (which looking at the ASD is not only due to seismic phase modulation of the IFO output field). Maybe some combination of this and the Marconi investigation would suggest why we have these forests of lines in the ASDs of the error signal?
   • Turn off the HEPAs on the PSL enclosure completely as a test, to see if that improves (i) phase noise due to air currents and (ii) mechanical pickup on the fiber producing  phase noise.

I tried all of these last night / overnight, here are my findings.

Analog locking of the homodyne phase:

See Attachment #1. 

• RF44_I was used as the error signal.
• The "C1:OMC-ZETA_IMON_OUT" channel is actually looking at the error signal monitor from the LB1005, and is uncalibrated in this plot.
• The "monitor" port on the demodulator board provides a convenient location for us to route the demodulated signal to an LB1005 box, while simultaneously digitizing both demodulated quadratures.
• Empirically, I found settings that could engage the lock. I also found that I couldn't increase the gain much more without destroying the lock. 
• The time domain signals look much "cleaner" in this analog feedback loop than when I achieved similar stabilization using the digital system. But I will quantify this more when I post some spectra of the in loop error signals.
• I will do some more characterization (loop TF measurement, error point spectrum in lock etc), but in summary, it looks like we still only have ~100 Hz UGF. So something in the loop is limiting the bandwidth. What could it be?
• The main problem is that the LB1005 isn't well suited to remote enabling/disabling of the lock, so this isn't such a great system.

Relative stability of two IFR2023s synchronized to the same FS725 Rb standard:

The electrical LO signal for demodulation of the 44 MHz photocurrent is provided by an IFR2023 signal generator. To maintain a fixed phase relation between this signal, and the phase modulation sidebands imprinted on the interferometer light via a separate IFO2023 signal generator, I synchronize both to the same Rb timing standard (a 10 MHz signal from the FS725 is sent to the rear panel frequency standard input on the IFR). We don't have a direct 44 MHz electrical signal available from the main IFO Marconi at the LSC rack (or anywhere else for that matter). So I decided to do this test at 55 MHz. 

• RF input of the demodulator was driven by 5*11.066209 MHz pickoff from the LSC rack.
• LO input of the demodulator was driven by 5*11.066209 MHz signal from the IFR2023 used for the RF44 demodulation setup.
• The outputs were monitored overnight. The RF44_Q channel had a DC level of nearly 0. So this channel is nearly a linear sensor of the phase noise between LO and RF signals.
• To convert ADC counts to radians, I offset the LO Marconi frequency by 100 Hz, and saw that the two quadratures showed pk-pk variation of ~24000cts. So, at the zero crossing, the conversion is 1/(24000/2) rad/ct ~83urad/ct.
• The result is shown in Attachment #2. The "measurement noise" trace corresponds to the RF. input of the demodulator being terminated to ground with a 50 ohm terminator.
• For comparison, I also overlay the phase noise estimate of an individual IFR from Rana. In his investigation, the claim is that the PLL that locks the IFR to the Rb timing standard has ~1kHz UGF, but if my measurement is correct, the relative stability between the two signal generators synchronized to the same timing standard already. degrades at ~1 Hz. Could be just a cts/rad calibration error I guess.
• In any case, we are far from saturating this limit in the homodyne phase lock.
• There are several sharp lines in this measurement too - but I don't know what exactly the source is. Of course the two marconis are plugged into separate power strips, so that may explain the 60 Hz lines and harmonics, but what about those that aren't a multiple of 60 Hz?

A look at the time domain signal:

With the Michelson locked on the dark fringe, the RF44 I and Q signals in the time domain are shown in Attachment #3 for a 1 minute stretch.

• The RF44 signal level bottoms out at ~40 cts. Okay, so this is the offset.
• However, the maximum value of the RF44 signal amplitude seems to be modulated in time. How can we explain this?

 I analyzed the data taken yesterday. 

The AS11 data in PRMI configuration is very bad, while the AS55 seems good enough:

The phase differences are 

AS11 = 21 +- 18 degrees (almost useless due to the large error)

AS55 = 11.0 +- 0.4 degrees

The AS55 phase difference is not the same measured in the last trial, but about half of it. The new length estimates are:

AS11 = 3.2 +- 2.8 cm

AS55 = 0.47 +- 0.01 cm

We can probably forget about the AS11 measurement, but the AS55 result is different from the previous estimate... Maybe this is due to the fact that Eric adjusted the PRCL offset, but then we're going in the wrong direction....

Summary:

I managed to achieve a few transitions of control of the XARM length using the ALS error signal. The lock is sort of stable, but there are frequent "glitches" in the TRX level. Needs more noise hunting, but if the YARM ALS is also "good enough", I think we'd be well placed to try PRMI/DRMI locking with the arms held off resonance (while variable finesse remains an alternative).

Details:

Attachment #1: One example of a lock stretch. 

Attachment #2: ASD of the frequency noise witnessed by POX with the arm controlled by ALS. The observed RMS of ~30pm is ~3-4 times higher than the best performance I have seen, which makes me question if the calibration is off. To be checked...

Summary:

1. The two arm lengths can be controlled reliably in the CARM/DARM basis using ALS error signals.
2. With a CARM offset to keep the arm cavitites off resonance, the PRMI can be locked using 3f error signals.
3. On attempting to reduce the CARM offset, I see a drop in the POP22 buildup in the PRC (correlated with the arm powers increasing). Not entirely clear why this is happening.

I ran some sensing measurements at various CARM offsets to check if the PRCL-->REFL33 and MICH-->REFL165 signals were being rotated out of the sensing quadrature as I lowered the CARM offset - there was no evidence of this happening. See Attachment #2. Other possibilities:

• CARM offset dependant offsets in the MICH/PRCL error points?
• Check the RAM due to the EOM? Perhaps the pointing / polarization control into the EOM got degraded.
• Angular stability of the PRC is still pretty poor, getting the angular feedforward back up and running would help the duty cycle enormously.

The IMC went into some crazy state so I'm calling it for the night, need to think about what could be happening and take a closer look at more signals during the CARM offset reduction period for some clues...

I looked at some signals for a 10 second period when the PRMI was locked with at some CARM offset, just before the PRMI lost lock, to see if there are any clues. I don't see any obvious signatures in this set of signals - if anything, the PRM is picking up some pitch offset, this is seen both at the Oplev error point and also in the POP QPD spot position. But why should this be happening as I reduce the CARM offset? The arm transmission is only ~5, so it's hard to imagine that the radiation pressure is somehow torquing the PRM. There are no angular feedback loops actuating on the PRM in this state except the local damping and Oplev loops.

The 1f signals are also changing their mean DC offset values, which may be a signature of a changing offset in the 3f MICH and PRCL error points? The MICH error signal is pretty noisy (maybe I can turn on some LPF to clean this up a bit), but I don't see any DC drift in the PRCL control signal.

I set up a photodiode (PDA10CF) in the IFO REFL beampath and the Agilent NA is sitting on the east side of the PSL enclosure. This was meant to be just a first look, maybe the PDA10CF isn't suitable for this measurement. The measurement condition was - PRM aligned so we have a REFL beam (DC level = 8.4V measured with High-Z). Both ITMs and ETMs were macroscopically misaligned so that there isn't any cavity effects to consider. I collected noise around 11 and 55 MHz, and also a dark measurement, plots to follow. The optics were re-aligned to the nominal config but I left the NA on the east side of the PSL enclosure for now, in anticipation of us maybe wanting to tune something while minimizing a peak.

Attachment #1: Results of a coarse sweep from 5 MHz to 100 MHz. The broadband RIN level is not resolvable above the dark noise of the photodiode, but the peaks at the modulation frequencies (11 MHz, 55 MHz and 29.5 MHz) are clearly visible. Not sure what is the peak at ~44 MHz or 66 MHz. Come to think of it, why is the 29.5 MHz peak so prominent? The IMC cavity pole is ~4kHz so shouldn't the 29.5 MHz be attenuated by 80dB in transmission through the cavity?

Attachment #2: Zoomed in spectra with finer IF bandwidth around the RF modualtion frequencies. From this first measurement, it seems like the RIN/rad level is ~10^5, which I vaguely remember from discussions being the level which is best achieved in practise in the 40m in the past.

Tried a bunch of things tonight.

1. Modified the "ELP300" filter module in the MICH filter bank - this was really a 4th order elliptic low pass with corner at 80 Hz, which was much too low. I tried upping the corner to 500 Hz, and reducing the order, while I was able to enable the filter, there was clearly a gain-peaking feature visible after engaging this module, so the exercise of reducing the high frequency MICH actuation requires more careful (daytime) loop optimization.
2. Tried adding some POPDC to the MICH/PRCL trigger once the PRMI was locked - I thought this would help if the problem was just with POP22 triggering turning off the MICH/PRCL loops, but the problem seems to persist with the mixed matrix trigger as well, once I reach a CARM offset where the arm powers exceed ~10, the PRMI loses lock.
3. One strange feature I don't understand is that with the PRMI locked with the carrier field resonant, when running the dither alignment servo to minimize REFLDC (= more carrier coupled into the PRC), the POPDC level also goes down, but TRX and TRY go up slightly. I confirmed that the beam isn't falling off the POP22 photodiode (Thorlabs PDA10CF), but I don't understand why these two DC powers should fall simultaneously - if I couple more carrier into the PRC, shouldn't the POPDC level also increase?

One possibility is that the arm buildup is exerting some torque on the ITMs, which can also change the PRC cavity axis - as the buildup increases, the dominant component of the circulating field in the PRC comes from the leakage from the overcoupled arm cavity. We used to DC couple the ITM Oplev servos when locking the PRMI. The TRX level of 1 corresponds to ~5W of circulating power in the arm cavity, and the static radiation pressure force due to this circulating power is ~30 nN, rising up to 300nN as the TRX level hits 10. So for 1mm offset of the spot position on the ITM, we'd still only exert 300 pN m of torque. I don't see any transient in the Oplev error signals when locking the arm cavity as usual with POX/POY, but on timescales of several seconds, the Oplev error point shows ~3-5 urad of variation.

Some short notes, more details tomorrow.

1. I was able to make it to CARM on RF only ~10 times tonight.
2. Highest stable circulating power was ~200 (recycling gain ~10) but the control scheme is still not finalized in terms of offsets etc.
3. DARM to RF transition was never fully engaged - I got to a point where the ALS gain was reduced to <half its nominal value, but IMC always lost lock.
4. CARM loop UGF of ~5 kHz was realized. I was also able to turn on a regular boost. But couldn't push the gain up much more than this. Should probably modify the boosts on this board, their corner frequencies are pretty high.
5. The increased FSS flakiness post c1psl upgrade is definitely hurting this effort, there are periods of ~20-30mins when the IMC just wont lock.

Attachment #1 shows time series of some signals, from the time I ramp of ALS CARM control to a lockloss. With this limited set of signals, I don't see any clear indication of the cause of lockloss, but I was never able to keep the lock going for > a couple of mins.

Attachment #2 shows the CARM OLTF. Compared to last week, I was able to get the UGF a little higher. This particular measurement doesn't show it, but I was also able to engage the regular boost. I did a zeroth order test looking at the CM_SLOW input to make sure that I wasn't increasing the gain so much that the ADC was getting saturated. However, I did notice that the pk-to-pk error signal in this locked, 5kHz UGF state was still ~1000 cts, which seems large?

Attachment #3 shows the DTT measurement of the relative gains of DARM A and B paths. This measurement was taken when the DARM_A gain was 1, and DARM_B gain was 0.015. On the basis of this measurement, DARM_B (=AS55) sees the excitation injected 16dB above the ALS signal, and so the gain of the DARM_B path should be ~0.16 for the same UGF. But I was never able to get the DARM_B gain above 0.02 without breaking the lock (admittedly the lockloss may have been due to something else).

Attachment #4 shows a zoomed in version of Attachment #1 around the time when the lock was lost. Maybe POP_YAW experienced too large an excursion?

Some other misc points:

• It was much quicker to acquire the PRMI lock with CARM held off resonance using the 1f signals rather than 3f - so I did that and then once the lock is acquired, transfer control to 3f signals (using CDS ramptime) before zeroing the CARM offset.
• The whole process is pretty speedy - it takes <5mins to get to the CARM on RF only stage provided the PRMI lock doesn't take too long (the transition from POX/POY to ALS sequence takes <1min).
• I am wondering what the correct way to set the offsets for the 3f error signals is? 
• The arm buildup is strongly dependent on the DC alignment of the PRMI - the best buildups I got were when I tweaked the BS alignment after the CARM offset was zeroed.

Yesterday I plugged the geophone and accelerometer output into the ADC, rather than the SR785, so I could collect for longer and take more data at once.

As per Rana's suggestion, I am also now taking the geophone output after the first op-amp in the circuitry following the geophone (a low-noise op-amp, OPA227). It acts as a buffer so I'm not just measuring other local noise sources (which explains why the geophone noise curve sort of matched the SR785 noise curve in my old plots).

With these changes, I remeasured the accelerometer and geophone noises as well as collected an ASD of a geophone sitting on the STACIS in open loop operation. I also looked up the noise specs for the various op-amps in the geophone pre-amp and high voltage board; everything I found, I added in quadrature to come up with an approximate op-amp noise value for the STACIS. All of this is plotted below:

I left the y-axis in V/rtHz instead of converting it to m/s/rtHz so that the op-amp noise could be compared to the other noises. All sensor data was taken with the sensors horizontal (noise data taken in granite and foam).

The accelerometer and geophone noise still appear to be similar, and the op-amp noise, at least according to specs, is low compared to the other noises. This implies there's not much to gain from switching the geophones with accelerometers nor with swapping out the op-amps for lower-noise components (unless the ones I couldn't find specs for were high-noise, though it seems like mainly low-noise components were used). 

Looks like you're just measuring the ADC noise. You should add ADC noise to your plot. To compare the geophones with the accelerometers, you have to correct for the preamp gain and plot them both in the same units.

To get above the ADC noise you can use an SR560 preamp. (AC Coupled, G = 100)

Based on Rana's comment I have gone through and moved all of the corner frequencies for the high pass filters in the SUS damping controllers to 30 Hz.  I did this for all optics (MC1, MC1, MC3, BS, ITMX, ITMY, PRM, SRM, ETMX, ETMY) all degrees of freedom (POS, PIT, YAW, SIDE).

Rana also suggested I turn off all of the BounceRoll filters until we get a chance to tune those individually for all the suspensions.

Finally, I normalized the MC SUSXXX filter banks to look just like all the other suspensions.

All damping filter banks for all degrees of freedom for all single suspensions should all be the same now (modulo the differences in the BounceRoll filters, which are now turned off).

This is getting closer, but with the whitening left OFF and the cts2um filter also OFF, none of the suspensions are working correctly. I'm shutting down all the watchdogs until someone gets around to setting the damping gains and filters correctly.

I'm attaching a screenshot of some of the problems I see so far with MC3.

I'm going to try to get the MC suspensions working OK for tonight so that we can use them for the PRMI locking work.

Update #1: None of the MC SUS DAQ channels are found by dataviewer....SUS debugging speed reduced by 10x.  Tue Jul 05 21:38:17 2011

Update #2: POS/PIT/YAW BIAS sliders now seem to work, but are ~1000x too weak to do anything.   Tue Jul 05 21:41:38 2011

I tried to fix all of the problems that I could identify in this screen shot:

• Fixed the TO_COIL output filter matrix screen to correctly point to the matrix element filter screens (all SUS)
• Removed MCL sections from SUS_XXX_POSITION screens, except for MC2.  I also modified the _POSITION screens for the ETMs to refer to ALS instead of MCL.
• Zeroed out all of the lockin gains in the TO_COIL matrices (MC SUS)
• Made sure all whitening filter were ON (all SUS)
• Made sure all cts2um calibration filters were on (all SUS)
• Made sure all oplev servos were on (all SUS)

Summary:

1. With the Michelson locked on a dark fringe, the f2-f1 signal at ~44 MHz does not seem to ever vanish, it seems to bottom out at ~2mV DC. Is this just an electronics offset? Not sure of the implications on using this as a locking signal for the homodyne phase yet.
2. The inferred relative phase fluctuations between the LO and RF fields using this 44 MHz signal is consistent with that from previous tests.
3. The laying out of the new, shorter, fiber patch cable seems to have helped to reduce the phase drift over minute time scales.
4. So far, I have not had any success in using the 44 MHz signal to close a servo loop and keep the homodyne phase locked for more than a few seconds at a time, and even then, the loop shape is sub-optimal as the in-loop error signal is not clean. Maybe some systematic loop shaping will help, but I think the dynamic range requirement on the actuator is too high, and I'm not sure what to make of the fact that the error signal does not vanish.

Details:

Attachment #1 shows the optical setup currently being used to send the LO field with RF sidebands on it to the air BHD setup.

• You can find a video of the large power fluctuations mentioned in my previous elog here. After tightening the collimator in the mount, the arrangement is still rather sensitive, but at least I was able to see some light on the DCPD on the AS table, at which point I could use this signal and tweak the alignment to maximize it.
• It is well known that the input beam to the IMC drifts during the day, either due to temperature fluctuations / PMC PZT stroke L2A / some other reason (see Attachment #4 for the power drift over ~12 hours, it is not monotonic with temperature). The fact that our collimating setup is so sensitive to the input pointing isn't ideal, but I noticed the power had only degraded by ~5% today compare to yesterday, so maybe the occassional touch up is all that is required.

Attachment #2 shows spectra of the relative phase drift between LO and IFO output field (from the Dark Michelson). 

• I still haven't overlaid a seismic model. There was some discussion about the TTs having a 1/f roll-off as opposed to 1/f^2, I don't know if there was any characterization at the time of installation, but this SURF report seems to suggest that it should in fact be 1/f^2 because the passive eddy current dampers are mounted to the main suspension cage on springs rather than being rigidly attached. 
• The noise at ~100 Hz is ~x2 higher if the spectra is collected during the daytime, when the seismic activity is high. Although this shouldn't really matter at 100 Hz? 
• There are also huge power-line harmonics - I suspect these are making it difficult to close a feedback loop, as I couldn't add a 60 Hz comb which doesn't affect the loop stability for a UGF of ~30-50 Hz. But if they aren't notched out, the control signal RMS is dominated by these frequencies.

Attachment #3 shows the signal magntiude of the signals used to make the spectra in Attachment #2, during the observation time (10 minutes) with which the spectra were computed. The dashed vertical lines denote the 1%, 50% and 99% quantiles.

• Koji asked me about the 55 MHz signal and why it doesn't vanish - for the dark Michelson, where the ITMs don't apply any relative phase on reflection to the carrier and RF sideband fields, we expect that the upper and lower sidebands cancel, and so there should be no intensity modulation at 55 MHz (just like we don't expect any for a pure phase modulated light field incident on a photodiode).
• However, from the I/Q demodulated data that is collected, it would appear that while the size of the signal does vary, it doesn't ever completely vanish. This implies some asymmetry in the sidebands (or at least, the transmission of the sidebands by the Michelson). I didn't estimate the effect of the Schnupp asymmetry, or if this asymmetry is coming from elsewhere, but the point is that for the conclusions drawn from Attachment #2 remain valid even though both the amplitude and phase of the 55 MHz signal is changing. 
• I also plot the corresponding histogram for the 44 MHz signal. You can see that it never goes to 0 (once I fix the x-label ticks). I don't know if this is consistent with some electronics offset.

Attempts to close a feeddback loop to control the homodyne phase:

• A digital PLL (a.k.a. Phase Tracker) servo was used to keep the demodulated 44 MHz signal in one (demodulated) quadrature, which can then be used as an error signal.
• Unlike the ALS case, the quantity to be servoed to 0 is the magnitude of the 44 MHz signal, and not its phase, so that's how I've set up the RTCDS model.
• I played around with the loop shape to try and achieve a stable lock by actuating on the PZT mounted mirror in the LO path - however, I've not yet had any success so far.

Summary:

After more trials, I think the phase tracker part used to provide the error signal for this scheme needs some modification for this servo to work.

Details:

Attachment #1 shows a block diagram of the control scheme.

I was using the "standard" phase tracker part used in our ALS model - but unlike the ALS case, the magnitude of the RF signal is squished to (nearly) zero by the servo. But the phase tracker, which is responsible for keeping the error signal in one (demodulated) quadrature (since our servo is a SISO system) has a UGF that is dependent on the magnitude of the RF signal. So, I think what is happening here is that the "plant" we are trying to control is substantially different in the acquisition phase (where the RF signal magnitude is large) and once the lock is enabled (where the RF signal magnitude becomes comparitively tiny).

I believe this can be fixed by dynamically normalizing the gain of the digital phase tracking loop by the magnitude of the signal = sqrt(I^2 + Q^2). I have made a modified CDS block that I think will do the job but am opting against a model reboot tonight - I will try this in the daytime tomorrow. 

I'm also wondering how to confirm that the loop is doing something good - any ideas for an out-of-loop monitor? I suppose I could use the DCPD - once the homodyne phase loop is successfully engaged, I should be able to drive a line in MICH and check for drift by comparing line heights in the DCPD signal and RF signal. This will requrie some modification of the wiring arrangement at 1Y2 but shouldn't be too difficult...

The HEPAs, on the PSL table and near ITMY, were dialled down  / turned off respectively, at ~8pm at the start of this work. They will be returned to their previous states before I leave the lab tonight.

[Koji, Steve, Kiwamu, Alberto]

- This afternoon we installed a few new optics on the BS table: GR_PBS, GRY_SM2, GRY_SM1.

- We pulled up the cables so that we had more freedom to move one of the cable towers farther South.

- Then we re-leveled the table. PRM OSEMs were adjusted to be nominal insertions.

- Koji released the earthquake stops on BS but the readout of the OSEMs was apparently frozen on the MEDM screens.
Initially we thought it was a software problem. a nuclear reboot didn't solve it. We spent the following three hours investigating the cause.
Eventually it turned out that the earthquake stops on BS weren't actually fully released.

We opened the tank and accessed to BS. Releasing the earthquake stops in full solved the issue. The OSEMs readout went back to normal.

July 29 Thu [Steve, Alberto, Kiwamu, Koji]

We placed some optics in the BS chamber.
The chambers are ready to be pumped down on Friday once the heavy door is placed.

- Clean room work

• Engraved two Y2 mirrors and PBS@532nm
• Engraved three DLC mounts
• Each of the mounts needs a 3.5 inch post. We found there is no stock of the post in the lab! Also the clamps!
• Took the posts from the temporarily removed optics although we need to return those optics into the table during the next vent.
• We should count the # of the mounts and count the needed posts. Posts and clamps can be either a DLC thick post or New Focus pedestal.

- In the chamber

• The terminal holder was moved as Alberto described
• The green steering optics were placed as Alberto described
• Note: the PBS is flipped in the mount (reflection side is back side)
• The table leveling
• Releasing EQ stops / Check OSEMs / Adjust OSEMs (BS OSEMs are untouched)

- After closing the chamber

• The BS OSEM mumbo-jumbo

{EricQ, Jenne]

More photos were taken.  Will post Monday, because too hungry now.

 Have eaten.  Here's a PDF with all the pictures to-date, along with a few notes.

Also, the first thing we did on Saturday was to fix the yaw pointing of MMT1, so that the beam hit the center of MMT2.  Then we had to touch PZT2 to compensate.  We put the iris target on the BS, and adjusted PZT2 until the beam went nicely through there.  The resulting beam looks good on the SRM, and teh beam is still hitting the AS camera.

I'm back to terrorize the PSL table again. The pictures I took yesterday were rubbish--today I'm using a clamp that Steve was nice enough to loan me. I'm starting now, at 10:09 am.

Summary:

1. It seems that after a x10 increase in the coil driver resistance, we will have enough actuation range to control (anti de-whitened) DARM without saturating the DAC.
2. The Barry puck doesn't seem to help us much in reducing the required RMS for DARM control. If this calculation is to be believed, it actually makes the RMS actuation a little bit higher.

See Attachment #1 for the projected control signal ASDs. The main assumption in the above is that all other control loops can be low-passed sufficiently such that even with anti-dewhitening, we won't run into saturation issues.

DARM control loop:

• I'm now calculating the DARM control signal in counts after factoring into account a digital DARM control loop.
• The loop shape is what we used when the DRFPMI was locked in Oct 2015.
• I scaled the overall OLTF gain to have a UGF around 200Hz.
• The breakdown of how the DARM loop is constructed is shown in Attachment #2.

De-whitening and Anti-De-whitening:

• The existing DW shape in the ITM and ETM signal chains has ~80dB attenuation around 100 Hz.
• Assuming ~5uV/rtHz noise from the DAC, 60dB of low-passing gets us to 5nV/rtHz. With 4.3kohm series resistance, this amounts to ~1pA/rtHz current noise (compared to ~3pA/rtHz from the Johnson noise of the series resistance). Actually, I measured the DAC noise to be more like ~700nV/rtHz at 100 Hz, so the current noise contribution is only 0.16pA/rtHz.
• This amounts to getting rid of the passive filter at the end of the chain in the de-whitening board.
• Attachment #3 shows the existing and proposed filter shapes.

It remains to add the control signals for Oplev, local damping, and ASC to make sure we have sufficient headroom, but given that current projections are predicting using up only ~1000cts of the ~23000cts (RMS) available from the DAC, I think it is likely we won't run into saturations. Need to also figure out what the implication of the reduced actuation range will be on handling the locking transient.

I think "OLG" trace is not labeled right; it would be good to see the actual OLG in addition to whatever that trace actually is.

Based on the first plot, however, my conclusion is that:

1. we don't need the passive isolator to reduce the control signal; the control signal is dominated by f < 10 Hz.
2. we should still look into isolators for the reduction of the f > 50 Hz stuff, just to make the overall DARM sensitivity better. But this does not have to be pneumatic since we no longer need 10 Hz isolation. It can instead be a solid piece of rubber to give us a ~20-30 Hz resonance. That would still give us a factor of 5-10 improvement above 100 Hz.
3. In this case, we only need a mass estimate of the end chamber contents with an accuracy of ~25%. If we think we have that already, we don't need to keep doing the jacks-strain gauge adventure.

I was a bit hasty in posting the earlier plots. In the earlier plot, the "OLG" trace was OLG * anti dewhitening as Rana pointed out.

Here are the updated ones, and a cartoon (Attachment #5) of the loop topology I assumed. I've excluded things like violin filters, AA/AI etc. The overall gain scaling I mentioned in the previous elog amounts to changing the optical sensing response in this cartoon. I now also show the DARM suppression (Attachment #4) for this OLG and the DARM linewidths for RSE. I don't think the conclusions change.

Note that for Signal Recycling, which is what Kevin tells us we need to do, there is a DARM pole at ~150 Hz. I assume we will cancel this in the digital controller and so can achieve a similar OLG shape. This would modify the control signal spectrum a little around 150Hz. But for a UGF on the loop of ~150 Hz, we should still be able to roll-off the control signal at high frequencies and so the RMS shouldn't be dramatically affected.

Steve is looking into acquiring 4.5kohm Vishay Wirewound resistors with 1% tolerance. Plan is to install two in parallel (so that we get 2kohm effective resistance) and then snip off one once we are convinced we won't have any actuation range issues. Do these look okay? They're ~$1.50ea on mouser assuming we get 100. Do we need the non-inductive winding?

Good question! I've never calculated what the resonance frequency would be if had an inductive resistor with our cable capacitance (~50 pF/m I guess).

I connected to the serial port using screen (through Terminal) and using Arduino's serial monitor and basically received the same strings that were received through python, so it's not a python issue. Checked the other TC 200 module and was also receiving nonsense, but it was all question marks instead of mostly K's and ['s.

This rules out a few possible reasons for the weird data. Next steps are to set up and configure the Raspberry Pi (which has been interfaced before) and see if the problem continues.

I acquired several spare OSEMs (in unknown condition) from Paco. They are stored alongside the shipment from UF.

[Jenne, Koji, Rana]

Thanks to turning off the AS55 analog whitening as well as the 1k:6k lead filter that Koji put into Darm's FM7, the DARM transition was more stable early in the evening.

The AS55 gain and offset did not change noticeably when we switched the AA on or off (switching happened while *not* using AS for any feedback).  Earlier in the evening, we did also check what happened with PRMI and REFL33 AA on vs. off, and REFL33 did have a many tens of counts offset on both the I and Q input channels.  I have turned the AA filters back on, but run LSCoffsets before trying to lock.

I'm not sure what was up, but somehow I couldn't lock the PRMI for about half an hour or so.  Very frustrating.  Eventually after futzing around, I was able to get it to lock with REFL33 in PRMI-only, and after that it worked again in PRFPMI with REFL165. 

With FSS slow around 0.5, MC has been a bit fussy the last hour.  Also frustrating.

Later on in the evening, I started taking out a bunch of the "sleep" commands from the up script, and many of the "press enter to continue" spots, but I think it might be moving too fast.  That, or I'm just not catching where I have too much gain.  Anyhow, near the middle/end of the CARM transition I am getting severe gain peaking at several hundred Hz.  I think I need to use a lower final gain.

So, progress on DARM, but maybe a little more fine-tuning of CARM needed.

Here's a DARM loop measurement, taken after both CARM and DARM were RF-only:

DARM_RF-only_13Mar2015.pdf

I have modified the DARM model from elog 11133, to include the fact that these are digital filters.

I have also extracted the data from elog 11143, and it together with the model.

The modeled loop has an arbitrary gain factor, to make it have the same 234Hz UGF as the measured data.

The modeled loop includes:

• Actuators
  • Pendulum (1Hz, Q of 4)
  • Violin filters
    • ETMs 1st, 2nd, 3rd order
    • MC2 1st, 2nd order
  • 3 16kHz delays for computation on the rfm model, transfer to the end sus models, and computation on end sus models.
  • Digital anti-imaging to get up to the IOP model
  • Delay of 64kHz for computation on IOP model
  • Analog anti-imaging
• Plant
  • Single 4.5kHz pole
• Sensor
  • Analog anti-aliasing
  • 64kHz delay for computation on IOP model
  • Digital anti-aliasing to get to LSC model rate
• Loop Shape (digital filters extracted from Foton file using FotonFilter.m)
  • DARM B's integrator (FM7)
  • DARM's low freq boost (FM3)
  • DARM's locking filter (FM5)
  • DARM's bounceRoll filter (FM6)
  • DARM's new lead filter (FM7)
  • Delay of 16kHz for computation on LSC model (includes Dolphin hop to c1sus rfm model)

There is a 1.5 degree phase discrepancy at 100Hz, and an 11 degree phase discrepancy at 900Hz, but other than that, the modeled and measured loops match pretty well.

For the measured frequencies, here are the residuals:

[Jenne Koji]

- Today the spots were moving more than the usual. The OPLEV screens showed that the spots are too much off from the center.

- Each vertex OPLEVs were checked and OPLEV wonderland was discovered: Other than the usual misalignment of the spots,
it was found that PRM/ITMX/ITMY beams were clipped somewhere in the paths, BS/PRM oplevs had many loose components
including the input lenses (they are still clamped by a single dog clamp THIS SHOULD BE FIXED ASAP).

- On the ITMY table there were so many stray optics. They were removed and put on the wagon next to the ITMY table.
THIS SHOULD BE CLEARED ON THE WEDNESDAY CLEANING SESSION.

- During this OPLEV session, LSCoffset nulling was run.

- After the OPLEV session, the locking became really instantaneous. We wonder which of the OPLEV cleaning, LSC offset nulling,
and the usual seismic activity decay in the evening was effective to make it better.

- Initially the lock was attempted with REFL33I/Q and some ~10sec lock streches were obtained. During this lock,
  the optical gain of AS55Q was measured in relative to REFL33Q. In deed they were calibrated to be the same
  gain at the input matrix.

- After the MICH signal source was switched to AS55Q, the lock streches became more regular and the minutes long.
We precisely tuned the phase of AS55 and REFL55 in terms of the differential excitation of ITMX/Y using lockin (FREQ 250, AMP 1000).

- We noticed that the AS port spot with AS55Q MICH was darker than the REFL33Q MICH. This suggests the existence of residual offset
in REFL33Q. In deed we observed +30cnt offset in REFL33Q when the PRMI is locked with AS55Q MICH.

- Phases and relative gains of the signals were as follows:

PRCL: REFL33I 1.00 =REFL55I +0.4
MICH: AS55Q 29deg x1.00 = REFL33Q -14deg x1.00 = REFL55Q 118deg 0.03?

- We tried to lock PRMI with AS55Q. The acquisition was not as easy as that with REFL33I. This might be from the saturation of the
REFL55I signal. This configuration should tested with different whitening gain. Handing off using the input matrix went well once the
lock was obtained by REFL33I.

- Handing off from AS55Q to REFL55Q was not successful.

- At the end of the session, Jenne told me that the POP PD still has a large diameter beam. (and a steering mirror with a peculiar reflection angle.)
==> THIS SHOULD BE FIXED ASAP because the normalization factor can be too much susceptible to the misalignment of the spot.

- The configuration of the filters:

PRCL FM3/4/5/6 G=+0.05 / NORM 0.04 POP110I
MICH FM4/5 G=-5.00 / NORM 0.01 POP110I (or none)

I locked each arm, and the Michelson last night, no problems after I increased the Yarm gain from 0.1 to 0.2 .  I checked the green beam alignment just before going home, and both of the green beams are locking on ~03 or 04 modes, so aligning them is on the list for today.

Summary:

I now think the excess noise in this circuit could be coming from the KEPCO switching power supply (in fact, the supplies are linear, and specd for a voltage ripple at the level of <0.002% of the output - this is pretty good I think, hard to find much better).

Details:

All component references are w.r.t. the schematic. For this test, I decided to stuff a fresh channel on the board, with new components, just to rule out some funky behavior of the channel I had already stuffed. I decoupled the HV amplifier stage and the Acromag DAC noise filtering stages by leaving R3 open. Then, I shorted the non-inverting input of the PA95 (i.e. TP3) to GND, with a jumper cable. Then I measured the noise at TP5, using the AC coupling pomona box (although in principle, there is no need for this as the DC voltage should be zero, but I opted to use it just in case). The characteristic bump in the spectra at ~100Hz-1kHz was still evident, see the bottom row of Attachment #1. The expected voltage noise in this configuration, according to my SPICE model, is ~10 nV/rtHz, see the analysis note.

As a second test, I decided to measure the voltage noise of the power supply - there isn't a convenient monitor point on the circuit to directly probe the +/- HV supply rails (I didn't want any exposed HV conductors on the PCB) - so I measured the voltage noise at the 3-pin connector supplying power to the 2U chassis (i.e. the circuit itself was disconnected for this measurement, I'm measuring the noise of the supply itself). The output is supposedly differential - so I used the SR785 input "Float" mode, and used the Pomona AC coupling box once again to block the large DC voltage and avoid damage to the SR785. The results are summarized in the top row of Attachment #1.

The shape of the spectra suggests to me that the power supply noise is polluting the output noise - Koji suggested measuring the coherence between the channels, I'll try and do this in a safe way but I'm hesitant to use hacky clips for the High Voltage. The PA95 datasheet says nothing about its PSRR, and seems like the Spice model doesn't include it either. It would seem that a PSRR of <60dB at 100 Hz would explain the excess noise seen in the output. Typically, for other Op-Amps, the PSRR falls off as 1/f. The CMRR (which is distinct from the PSRR) is spec'd at 98 dB at DC, and for other OpAmps, I've seen that the CMRR is typically higher than the PSRR. I'm trying to make a case here that it's not unreasonable if the PA95 has a PSRR <= 60dB @100 Hz.

So what are the possible coupling mechanisms and how can we mitigate it?

1. Use better power supply - I'm not sure how this spec of 10-50 uV/rtHz from the power supply lines up in the general scheme of things, is this already very good? Or can a linear power supply deliver better performance? Assuming the PSRR at 100 Hz is 60 dB and falls off as 1/f, we'd need a supply that is ~10x quieter at all frequencies if this is indeed the mechanism.
2. Better grounding? To deliver the bipolar voltage rails, I used two unipolar supplies. The outputs are supposedly floating, so I connected the "-" input of the +300 V supply to the "+" input of the -300 V supply. I think this is the right thing to do, but maybe this is somehow polluting the measurement?
3. Additional bypass capacitors? I use 0.1 uF, 700V DC ceramic capacitors as bypass capacitors close to the leads of the PA95, as is recommended in the datasheet. Can adding a 10uF capacitor in parallel provide better filtering? I'm not sure if one with compatible footprint and voltage rating is readily available, I'll look around.

What do the analog electronics experts think? I may be completely off the rails and imagining things here.

Update 2130: I measured the coherence between the positive supply rail and the output, under the same conditions (i.e. HV stage isolated, input shorted to ground). See Attachment #2 - the coherence does mirror the "bump" seen in the output voltage noise - but the coherence is. only 0.1,  even with 100 averages, suggesting the coupling is not directly linear - anyways, I think it's worth it to try adding some extra decoupling, I'm sourcing the HV 10uF capacitors now.

Yes. The datasheet has a recommendation circuit with 10uF caps. Companies are careful to show reproducible, reliably functional circuit examples on datasheets. So, if the caps are there you should try to replicate the design.

true. also try to choose a cap with a goow high frequency response. In the Electronics Noise book by Ott there's some graph about this. I bet you good do a Bing search and also find something more modern. Basically we want to make sure that the self resonance is not happening at low frequencies. Might be tought to find one with a good HF response, a high voltage rating, and > 1uF.

I'm still having issues with the STACIS oscillating uncontrollably with the slightest extra vibration, but with some more added weight both x and z direction are stable if you don't disturb the setup.

I took more transfer functions of the STACIS. In the last data I took Jenne pointed out that the geophone signals were not correlated well with the driving signal, so I increased the amplitude of the driving signal and am looking in x and y too instead of just z. 

Details of the driving signal: 25 mV, swept sine from 0.1 to 100 Hz from the SR785. 

NOTE: The data below was all transferred from the SR785 using netGPIB, which works fine, if anyone was interested in using it.

Open loop in the y direction, taken with the y geophone (magnitude on top, phase on bottom):

Open loop in the x direction, taken with the x geophone (with some extra weight to try to make the closed loop more stable):

Open loop in the x direction, taken with accelerometer instead of geophone: