All parts for the motion of the retaining rings have been received and are ok. We're going to clean and bake them. 

The autocenter script wasn't working in a very robust way (sometimes the socket connection to the controller failed, some times no motion was obtained after issuing a command). So I rewrote the interface to the Newport controller using the http server interface. The code is in picomotor8742_http.py. This version seems much more robust.

There is a large wandering line that spans all frequencies. Not sure what the origin is, but it will need some noise hunting. Here are spectrograms of all four QPD X signals. The line is visible in all of them, at the same frequency. So it's definitely something external coupling into the light or the QPD electronics. It's not visible in the sum of the QPD quadrants, but maybe just because it's buried in noise.

Finally, I tried to excite the disks, but I couldn't get any motion of the optical lever beams. To be investigated, there might simply be some cables disconnected.

My bad, a cable was disconnected.

The Thorlabs laser has been misbehaving for the whole weekend. Even after many days being continuosly on, the wandering line is still moving all over the frequencies.

So this morning I swapped in a JDSU 1125P borrowed from the 40m lab, which provides about 6.8 mW of power. I tested it over the weekend on a separate test table, and after one day or so of operation the power looks reasonably stable. Now it's been on for a few hours: there is still a line moving around, but it's slowing down and hopefully setting down in a good place.

I started a series of test measurements on the samples that were already installed.

• Quiet time before excitation: 1170443490
  Excitation (broad band) at 1170443523 (60 s)
  Quiet time after excitation: 1170443586

The high power lasers I tested so far (the Thorlabs 21mW and the JDSU 1125P) are noisy: they both have wandering lines that from time to time are alised down into the base band, destroing the measurement. 

I have three JDSU 1103P units: two of them dlived about 2.5 mW, the third one delivers about 1.4 mW. One of the 2.5mW was installed in the test setup. I swapped it out with the 1.4 mW, so now I have two good 2.5 mW laser. My plan is to modify the new setup to use those two lasers in parallel, splitting each one in two, for a total of four beams of about 1.2 mW each.

The new optical layout is atttached.

Today I swapped out the 8mW laser and installed two 2.5 mW lasers. I rebuilt the input part of the optical levers and re-aligned everything. See below for a picture of the new setup: red beams are input, yellow beams are output. I also installed a protective screen all around the table, to abvoid any suprios beam to get out.

The lasers are behaving well, there is no high noise or wandering lines. The spectrum below is taken in air: that explains the excess of noise in the few kHz region.

This afternoon I installed the picomotor and the translation stage that will be used to move the retaining rings up and down. No partciular problem: I only had to add some small aluminum foil shims between the ear of some rings and the square plate, to make the rings as horizontal as possible.

I tested the motion: with 300000 steps it's possible to move the rings all the way from the parked (down) position, to the up position. I also checked that when the rings are up, I can place four substarates and they fall properly into the alignment groove. Since the maximum speed of the picomotor is 2000 steps/s, it takes 150 seconds to move up and down the ring. 

Finally, positive steps means that the rings are moving up, negative that they're moving down.

I raeligned the optical levers to the position I obtained by centering the samples with the rings. I haven't tested the repeatability yet.

The ring motion up and down was not very smooth, again due to friction on the centering pins.

So, after centering the rings using the pins and securing the rings to the translation stage, I removed all pins.

Now the motion up and down is very smooth.

I still have to fine tune the amount of steps that are needed to go up and down.

However, initial tests don't show a good repeatability of the positioning. My main suspect is that the vibration caused by the picomotor cause the disks to slip on the silicon lens. Indeed, when the disks are sitting on the rings, one can clearly hear them "rattle".

I did two set of measurements with the new coated samples from Montreal. I reshuffled the position in the two measurements. In both cases, the measurement being performed in bay 4 was bad, in the sense that it was very hard to see excited modes. Since the two measurements were carried out with two different disks, it's clear it is a problem with that setup.

SOLVED: there was a connection problem for the DAC output signal controlling the switch

REALLY SOLVED: it was not a cabling issue. The power supply for the switching box had the current limiter on: when all four switches are closed, the box drain about 270mA, which is more than the limit of 250mA. Therefore the power supply voltage dropped and only three switches were actually closed. I switched the power supply to 500mA range and maxed the current limit. Now all four switches are working properly

Since I had recurrent problems with the picomotors used for QPD3, I swapped them with another Newport motorized mirror that was previously used in the Crackle1 experiment. This is the same model used for the other three QPD centering. Everything looks to be working fine now.

I also realigned all optical levers and swapped out an iris with a smaller one, to avoid beam clipping. All beam paths look clear now.

This afternoon I removed the old periscope from CR0 and installed a new one with finely adjustable mount, like those in the new chamber. I realigned the optical lever to the horizontal refererence.

This afternoon I installed the new Lumentum (former JDSU) HeNe laser, model 1103P in CR0.

Realigned everything.

I installed a sample in the chamber to reflect a beam back inot the QPD. Checking the QPD signals over a hour and more did not show any sign of excess noise or instability.

For the optical levers we are going to use the same QPD that are used in aLIGO optical levers (see T1600085 and D1100290): Hamamatsu S5981

Based on the aLIGO design, I put together a design fof the QPD boards, see the first attached PDF file. Some comments:

• I'm using a quad opamp to implement the transimpendance and a doubel stage of whitening for the readout of each quadrant. I haven't decided yet which is the right whitening
• The bias to the QPD is just coming from the +15V with a capacitor to ground, as in the aLIGO setup. However, I allowed the possibility of supplying a lower noise bias if needed, by removing R0 and connecting an external circuit
• There is no sum or difference computation, each quadrant is sent directly to the ADC with a differential driver (DRV135)
• Each board hosts one QPD, and it is interfaced with a 14 pin header (we're going to use twisted cables for the signals)

The stable power supply will be provided by an additional board, which will also interface 8 QPD boards to the ADC connector, see the second attached PDF. The ADC signal grounds can be connected directly to the power supply ground, left floating, or connected with a RC filter, depending on what we find to be the best solution.

Total cost estimated for PCB manufacturing and components, including the QPDs is less than 3k$, for a total of 12 boards (we need 8, plus some spares)

Koji, Rich, and I recently came up with a new QPD design which is better for general lab use than the aLIGO ones (which have a high-noise preamp copied from iLIGO). 

https://dcc.ligo.org/LIGO-D1500467

This page has the mechanical drawing only, but perhaps Rich can tell us if he's ready to make the first version for you or not. I think you can get by with the old design, but this new one should be lower noise for low light levels.

After a very useful discussion with Rich this morning, I think the circuit based on aLIGO optical levers design should be good for our applications.

It uses a LT1125 as input stage, which has

• (maximum) current noise of 1e-12 A/rHz @ 100 Hz and 0.4e-12 A/rHz @ 1 kHz
• (maximum) voltage noise of 4e-9 V/rHz @ >100 Hz

We expect to send about 5 mW into the disk, getting back a 4% reflection, which would correspond to 200 uW on the QPD. Let's say we lose half of this power in reflections through viewports and such, so we have a total of 100 uW on the QPD, or 25 uW on each quadrant. From the QPD datasheet the repsonse is about 0.4 A/W, so we have a photocurrent of 10 uA. The corresponding shot noise limit is about 1.8e-12 A/rHz.

Using a transimpendace of 200k, the noise at the output of the transimpendance is

• shot noise = 3.6e-7 V/rHz
• voltage noise = 4e-9 V/rHz
• current noise = 2e-7 V/rHz @ 100 Hz, 0.8e-7 V/rHz @ 1 kHz

So in the worst case the current noise will be about half of the shot noise. This seems good enough.

The circuit design sent out for fabrication is available in the DCC: D1600196

Here are some screenshots of the disk assembly and a look at how four of them will sit into the vacuum chamber. The Solidworks models are available here: D1600197

Attached a first layout of the optical lever systems. The beam spot radius on the QPD is about 0.8 mm, and the lever arm length is of the orer of 1.4-1.5 m for all four beams.

An improved design is attached. I modified the input telescope to avoid using shor focal length lenses, to make it less critical, and to reduce the beam spot radius at the QPD to 0.5 mm.

A preliminary design of the ESD board is available on the DCC: D1600214

I measured the beam profile of the new Thorlabs HeNe (21.8 mW measured). The beam waist is 355 microns, very close to the laser output port.

Using those numbers and the optical gain optimization algorithm, I tweaked the optical lever design. The simplest solution uses two lenses right after the laser to focus the beam down to about 300 microns on the QPD. The arm lever length is about 1.6 m, corresponding to an optical gain of about 18000/rad. I updated the DCC drawing in D1600213

Since my experiment with coil and magnets didn't work out very well, here's a new concept for the motion of the four retaining rings (all together) using a translation stage and a picomotor. This follows the same idea put forward by Steve Penn. The translation stage is a Newport 9066-COM-V and the picomotor (which we already have) is a Newport 8301-V. Both stage and picomotor are vacuum compatible (rated at 1e-6 Torr) and tested down to 1e-8 Torr by Steve.

Here's the jig integrated in the full system:

The PCBs for the QPD circuit and ADC interface are here and look ok. All electronics components are also here (except for the ADC connector which should be ordered separately from Mouser, after confirming that the ADC we're going to use have the same cable as the one we use in the Crackling Noise experiment). The QPD will be shipped on 06/17.

I moved the unused rack from the Crackling Noise lab to the C.Ri.Me lab. It will be used for the new cymac. I also started putting the new workstation together, but I'm missing some adaptors for the monitors.

So here's the final proof that the glitches I see are digital:

• There is a positive jump (of random size) every second, every time 0.45s after the beginning of the second. There is a negative jump at a less constant time, but between 0.65s and 0.8s after the beginning of each second
• I moved my whole setup to the crackle lab. This included: HeNe laser with power supply, QPD with cable, interface board and power supply. I don't see any glitch there

I swapped the ADC board with a second one in the new cymac, but no change: glitches are still there.

The old channel X3:CR1-PRESSURE_LOGTORR does not exist anymore. The new channel is now directly in torr and it is called X3:CR1-PRESSURE_TORR.

I had to write a C function to compute the 10^x operation, since it is not included in the RCG routines. Also it's not possible to include library functions, so I had to write an ad-hoc function, which first compute the integer part of the exponent, and then approximate the fractional part with linear interpolation and a look-up table. Code is in /opt/rtcds/userapps/release/models/pow10.c

11:25am LT: closed valve between roughing and turbo pumps, switched off both pumps. Turbo pump is slowing down

After lunch I opened the chamber and removed everything from the inside.

The chamber around the vacuum gauge is really dirty now, see picture:

In addition, the electrostatic driver shows some signs of "burn" even though it was still working quite well. Unfortunately, whatever happened contaminated our sample:

I ran a set of COMSOL simulations to determine the dependency of the frequency of each eigenmode on the disk thickness and diameter, within the tolerances. I chose wide ranges: diameter 75.0 +- 0.1 mm and thickness 1.0 +- 0.1 mm, much more than the expected tolerances. It turns out that the frequencies depends almost exactly linearly on both variables: mostly on the thickness and negligibly on the diameter. The following plots shows: the mode shape and frequency (left), the frequency dependency on the two variables (center), the residual of a linear fit and the functional form of the fit (right).

I'm including only the modes that will be measurable by our system (no motion in the center, frequency below 32kHz. Since the disks in my simulation is completely round, I'm showing only one mode for each doublet.

I installed a dielectric laser line filter (Thorlabs FLH05633-5, center wavelength 633 nm, FWHM 5 nm) in front of the QPD. In this way we are no more affected by the room light. In the plot below blue is without filter, red with filter. A lot of peaks at high frequency are eliminated by the filter.

The plot below shows the QPD signal quadrant signals in a few different configurations: blue with the room light on and laser off, red with the room light off and laser off, green with the laser on. With the filter installed, when the laser is on we are dominated by its intensity noise, which shows a lot of peaks at high frequency. Those peaks are not completely eliminated by the difference of the quadrants.

Here's how I imagine the mode search to proceed:

1. the user select the list of nominal frequencies and the frequency range for the search
2. the automation creates an excitation which is a sum of band-limited white noise, centered around each of the modes, with the desired bandwidth. The excitation is normalized to match a reasonable fraction of the DAC range
3. the excitation is injected for a given amount of time and each band is searched for the mode. In each band, if a mode is found, the frequency is estimated together with the maximum SNR obtained and the results are saved
4. if some modes are not found, the automation will try to improve the noise, by scaling all band amplitudes in such a way to have the same resulting SNR for the modes. In this way we should get more dynamical range and could increase the excitation for the modes that were not found
5. if some modes are still unidentified, then the user will be asked to choose between a few options:
   1. skip the mode
   2. try to excite it with an excitation focused on that band only
   3. increase the bandwidth for the search
   4. maintain the same bandwidth but move the band to adjacent frequencies
6. At the end all identified modes will be stored in memory and saved to a file defined by the user

Disk excited at 12:01pm. Exited the room at 12:03pm.

Opened the chamber at about 2:30pm, got the disk out for edge polishing, installed it back at 3:30pm, pumping down at 3:40pm.

Stopped the roughing pump at 4:44:00pm (+60 seconds clean data, GPS 1155944657). Switched on the HV amplifier, excitation at 4:47:30pm. Recentered QPD, clean data from 4:48:30pm (GPS 1155944927)

After a first look at the data, it seems that something went wrong. I restearted the roughing pump and will pump overnight.  I found the QPD miscentered, so I centered it again.

Excited again at about 5:46:35pm. Clean data from 1155948460

We initially received 20 disks (75 mm diameter, 1 mm thickness) from Mark Optics. Here's their status as of today

• MO-01 was annelaed and measured at LMA. It was then taken back to Mark Optics by Julie Houser at the time of her visit
• MO-02 was initially used to test the "dirty" measurement apparatus, then it was installed into the new chamber and measured on Saturday August 20th
  Then this disk was used as a guinea pig for the laser edge polishing, and it's now in a pretty bad shape
• MO-03 was installed into the new chamber and measured a couple of times on Sunday and Monday. Then we laser polished the edges, with good results, although there are small damages on the surface. It has been installed back into the chamber and a couple of rind-down measurement carried out on Monday and Tuesday.
• MO-04 is still untouched

All the other disks have been sent back to Mark Optics to grind out flats. 

Two good ring-downs measurements were performed on MO-02. The first one was already reported in a previous elog entry. I performed another measurement, and refined the mode identification. I think I had misidentified some modes in my previous analysis. The following plot shows the difference between the modes as predicted by COMSOL and as measured. A clean quadratic trend is visible and fitted:

Here's the spectrum with all the modes:

And the updated Q measurement plot:

A second ring down was measured on Monday morning . Here are the relevant plots:

This is the same disk as before, but almost all Q values are systematically higher. Here's a direct comparison:

I'm not sure what changed between the two measurements, except for a re-alignment of the QPD. The disk might have moved a bit...

Disk excited with white uniform noise, amplitude 5 V, for some tens of seconds.

Excitation off at

PDT: 2016-09-13 10:32:23.887615 PDT
UTC: 2016-09-13 17:32:23.887615 UTC
GPS: 1157823160.887615

At 1:05pm LT I stopped the roughing pump and started a ring-down measurement. Pump restarted at 2:18pm LT.

Excitation started at 20:15:30LT, 20 seconds long. The excitation is band-limited (10 Hz) centered around each of the predicted mode frequencies. Amplitude inversely proportional to the mode frequency.  The system was quiet before the excitation for many minutes.

For reference, here's the code used for the excitation:

from noise import *
from numpy import *
x = loadtxt('predicted_modes.txt')
bands = map(lambda x: [x-5,x+5], x)
ampl = x/x[0]
xx = multi_band_noise(bands, ampl, T=20, fs=65536)

n = AWGNoiseStream(1e-2*xx, channel='X3:CR1-ESD_EXC', rate=65536)
n.start()

The plot below shows three measurements of the Q of the same disk: during the first two the roughing pump was on, while during the third it was off. No significant difference is visible in the Q values.

Same as in elog #110, but now the amplitude is proportional to frequency squared:

ampl = (x/x[0])**2
xx = multi_band_noise(bands, ampl, T=20, fs=65536)
n = AWGNoiseStream(4e-4*xx, channel='X3:CR1-ESD_EXC', rate=65536)
n.start()

Noise stopped at 8:27:40am LT.

Turbo pump off and spinning down at 9:37am LT. Pumo completely stopped at 11:15am LT

Openend the chamber and removed the sample at ~11:20am LT

Turbo pump off and spinning down at 9:37am LT. Pumo completely stopped at 11:15am LT

Openend the chamber and removed the sample at ~11:20am LT

I made a COMSOL model that can compute the distribution of elastic energy for each mode, dividing it into:

• bulk and shear energies (integrated over the entire volume)
• edge energy (integrated over the edge surface only)
• surface energy (integrated over the top and bottom surfaces)

Then I used the measured Q values for the MO_101 disk and tried to see if I could reproduce it with the energy distribution. The first plot here shows that the loss angle of the disk (inverse of the Q) has a trend that is already quite well reproduced by the ratio on edge energy over total energy:

In particular the edge energy distribution is enough to explain the splitting of the modes in families. This fit is obtained assuming that the edge losses are uniform along the entire edge, and frequency independent. If we assume a "thickness" of the edge of the order of 1 micron, the loss angle is about 3.5e-3, which seems resonable to me since the edge is not polished.

Then I tried to improve the fit by adding also bulk, shear and surface losses. It turns out that shear is not very important, while bulk and surface are almost degenerate. The following plot shows a fit using only edge and surface losses:

The result is improved, expecially for the modes with lower loss angle. Again, assuming a surface thickness of 1 micron, the main surfaces have a loss angle of 1.3e-5, while the edge is 2.3e-3.

Including all possible losses gives a fit which is basically as good as the one above:

However, the parameters I got are a bit differentL: the surface losses are reduced to zero, while bulk dominates with a loss angle of 1.4e-4, and shear is not relevant. 

In conclusion, I think the only clear message is that the Q of our disks are indeed limited by the edge. The remaining differences are difficult to ascribe to a paritcular source. Since th disks are thin, I tend to ascribe them to the surface, which would imply that we are far from being able to see the bulk/shear losses. If I use only edge and surface losses, I found as expected that the polished main surfaces have much lower loss angle by a factor 200 or so.

Same plot as below, but this time with estimated 95% confidence intervals for the Q values, as obtained from the fit only.

As a test, I installed MO1 (the disk with the burn mark, used for the first edge laser polishing test) and started pumping down. Roughing pump on at 3:05pm, turbo pump on at 3:16pm.

The pressure is at abour 3e-6 Torr. I centered the QPD and started an excitation. The HV amplifier manual states that the driver can source both positive and negative voltage, so this time I didn't add any offset, but simply drove with 1000 V peak to peak. After the excitation the QPD was slightly miscentered in X and I had to manually recenter it.

Good data starting from

PDT: 2016-09-20 16:38:09.330642 PDT
UTC: 2016-09-20 23:38:09.330642 UTC
GPS: 1158449906.330642

NOTE: it's a good idea to take a look at both the X and Y signals for each mode. Some of them look stronger in Y than in X. So far I only used X.

New excitation (2000V) at about 8:06am. Had to recenter the QPD again after the excitation.

Engaged the 500Hz high pass filter on the ESD filter bank. New excitation ended at 8:11am. Amplitude 1000 V. Recentered the QPD at 8:11:35am

Just for fun, I installed the disk thas has been etched in the center with "1234". I figured out that the ESD PCB was probably too close to the disk, so I moved it a bit up.

Pump down started at about 2:38pm LT.

QPD centerd, quiet data, light off, one minute from

PDT: 2016-09-22 09:46:29.393609 PDT
UTC: 2016-09-22 16:46:29.393609 UTC
GPS: 1158598006.393609

Excitation (2kV) stopped at

PDT: 2016-09-22 09:49:03.784165 PDT
UTC: 2016-09-22 16:49:03.784165 UTC
GPS: 1158598160.784165

Some easy to read reviews on thin films and ideal glass for people getting started in this Q measuring game.

1. M. D. Ediger, in PNAS  (2014), pp. 11232–11233.

2. A. J. Leggett and D. C. Vural, arXiv cond-mat.dis-nn, arXiv:1310.3387 (2013).

3. L. Berthier and M. D. Ediger, arXiv cond-mat.mtrl-sci, 40 (2015). Phys. Today.

4. G. Parisi and F. Sciortino, Nature Materials 12, 94 (2013).

5. S. Singh, M. D. Ediger, and J. J. de Pablo, Nature Materials 12, 139 (2013).

Annealing run (447-448) on 3" wafers - Crime 10/27/2016 https://dcc.ligo.org/T1600485-v1

This morning I found the cymac and the workstation rebooted, so I suspected a power cut in the last days. However, the function generator and the power supply for the QPD were off. So somebody must have turned them off.

Please write those actions in the elog!