40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  Cryo Lab eLog, Page 33 of 58  Not logged in ELOG logo
ID Date Author Type Category Subject
  1278   Sat Jul 25 16:50:13 2015 DmitryDailyProgressSiFi - ringdownTaiwan Cantilever - Update

Yesterday Matthew and I opened the cryostat in order to reclamp the cantilever (hoping to get higher Q) and move the ESD to make it face the middle of the cantilever so that the second mode would be excited more effectively.

In attempt to compensate the angle between the upper and the lower parts of the clamp and make the pressure distribution more even, we removed guiding rods and replaced them with screws:

We tried ti clamp it with 4 screws using the central holes, but the holes in the upper and lower parts of the clamping did not match so we used two guiding rod holes instead. We also left about 1/10 of the thick part of the cantilever outside the clamping. This resulted in frequencies shift (105.5Hz for the first mode and 655Hz for the second) and bad quality factor:

Pressure, Torr Frequency, Hz Q
5e-3 105.5 5.3e3
5e-3 655 2.4e3
1.5hrs evac. time, gauge reading 1e-7 105.5 9.6e3
-------------------------------------//--------------------------- 655 3e3
13hrs evac. time, gauge reading 1e-7 105.5 1.5e4
---------------------------------//--------------------- 665 1.4e3

We failed to make continuous Q measurements for the 2nd mode, as we got almost no excitation, and the effectiveness of the ESD at the 1st mode also was poor.

Today we reclamped the cantilever (holding it now by almost the full length of the thick part - minimum clamping loss according to COMSOL, two front screws only. Using 500um Si shim we were able to compensate the angle between the upper and the lower part of the clamp). We also removed part of the ESD support to make it closer to the cantilever:


The quality factor of the first mode is worse than the one achieved with the first clamping (CRYO:1276), the one of the second mode - better. We were able to repeat continuous Q measurements for the second mode, got excitation for the second mode but the UGF is still too low.

Pressure, Torr Frequency, Hz Q Comments
1e-2 106 5e3 ringdown
1e-2 663.5 1.5e4 ringdown
1e-3 106 9e3 continuous
1e-3 663.5 2.1e4 ringdown
3hrs evac time (gauge shows 2e-7) 106 1.4e4 continuous, laser attenuated (attenuator's transmission - 25%)
-----------//--------- 663.5 3.0e4 ringdown, exponential fit, laser attenuated
-----------//--------- 106 1.4e4 continuous, full laser power
-----------//--------- 663.5 3.0e4 ringdown, exponential fit, full laser power
9hrs evac time 106 2.3e4 ringdown, exponential fit
----------//---------- 663.5 3.2e4 --------//-------------------------


  1277   Fri Jul 24 22:38:46 2015 DmitryNotesVacuumPfeiffer Vacuum Pump Conundrum

[Update: MSU pump -- problem solved by opening/closing gas ballast valve]

Symptoms: pump does not reach the full speed (1500Hz). It goes up to some level (between 1000Hz and 1300Hz), stops accelerating and the speed starts dropping after some time. 

Problems detected

Backing pump was set to operate at 230V - switched to 115V. Main problem remains.

Current condition (test performed on 07/24/15):

Setup: pump - KF40-KF25 reducer - KF25 hose - KF25 block off flange. Pump starts ok, after ca. 4 minutes the rotation speed reaches 996Hz (Driving Current: 3.31A, Bearing Temp: 29oC). After 5.5 minutes since startup the rotation speed starts dropping. Driving current level stays the same. At this point I switched it off, bearing temperature was 31oC. It came to almost complete stop in about 4 minutes. I checked the other pump's bearing temp - it was 30oafter 2 hours of non-stop operation at full speed.

Possible solutions suggested by Pfeiffer Vacuum's manual:

a) Check backing pump.

b) Open and close gas ballast valve on the backing pump to release possible condensate (located the valve, failed to open it)

c) Check/replace oil reservoir.

d) Check/replace bearing



a) Bearing is one of the possible culprits - it may be useful to compare bearing temperature readings to the ones reported by the other properly working pump as faulty bearing should result in higher temp.

b) The same symptoms were observed with the Pfeiffer Vacuum pump at MSU. The pump is known to have worked properly in the past. Changed the oil reservoir - did not solve the problem.

Update: MSU pump problem solved by opening/closing gas ballast valve. Procedure: open the gas ballast valve, switch on the pump, let it reach the final rotation speed, close the gas ballast valve.

  1276   Thu Jul 23 23:40:51 2015 DmitryDailyProgressSiFi - ringdownTaiwan cantilever - Ringdown - Continuous Q-measurements

[Nic, Zach, Matthew, Dmitry]


On Tuesday we opened the cryostat and switched the Painter 2 cantilever for the Taiwanese one (CRYO:1211). ESD seemed to be a bit too high above the cantilever and was turned upside down (the cantilever was facing the wrong side of the ESD). Nevertheless it provided enough force to excite both the first mode (106Hz) and the second one (663Hz).

Quality Factor (1st variant)

Date Frequency, Hz Q Pressure, Torr Technique
07/21/15 106 (7-8)*103 10-2 continuous Q
07/21/15 106 9*103 10-3 continuous Q
07/21/15 633 1.3*104 10-3 ringdown
07/22/15 106 3*104 10-(4-5) (*) continuous Q
07/22/15 663 1.6*104 10-(4-5) (*) ringdown

(*) it is an estimate, the gauge actually showed 10-7 torr


On Wednesday we opened the cryostat again, cleaned the surface of the cantilever (there were several smears on both surfaces), lowered the ESD (so that it is now closer to the cantilever and the side with electrodes is facing the cantilever – see the photo below) and Zach installed the guiding rods for the clamp.

Quality Factors (2nd variant)
Date Frequency, Hz Q Pressure, Torr Technique
07/22/15 106 7*103 10-3 (*) ringdown
07/22/15 663 (9-10)*103 10-3 (*) ringdown
07/23/15 106 2.3*104 10-5 (**) ringdown
07/23/15 663 8*103 10-5 (**) ringdown

(*) it is an estimate; evacuation time - ca 1.5 hours, the gauge already showed 10-7 torr.

(**) also an estimage - evacuation time - about 12 hours.

Either the surface became actually dirtier or more dusty, or the previous clamping was better.


Today with Nic's help Matthew and I were getting ourselves familiar with continuous Q measurements technique. We were able to repeat continuous Q measurement for the first mode and got close to setting up one for the second mode. The ESD is close to the node of the second mode, so the excitation is not as effective as for the first one, and the frequency is higher so it is difficult to get UGF high enough.

Nevertheless we were able to perform “proof of concept” experiment where we were driving both frequencies simultaneously and were able to change amplitudes of both modes independently.

After that we once more estimated the current quality factor values:

160Hz  - 3.2*104 

663Hz -  7*103


I made a COMSOL model of the cantilever and the clamping. For some reason (most probably the big aspect ratio of cantilever's length to it's thickness) COMSOL's results are mesh dependent, no matter how coarse/fine the meash is. 3D model failed, but 2D one gives consistent results using specially designed mesh (it predicts correct 106 and 663 Hz frequencies and correct quality factors given some bulk loss factor value). Using this model I calculated the dependence of the clamping loss (through energy leakage) on the length of the part of the cantilever that is clamped (L=0..10mm). COMSOL predicts minimum loss if almost all of the cantilever's thick part is clamped  L(min)~9.7mm.

  1275   Thu Jul 23 19:19:33 2015 BrianUpdateSimulationScripts for cavity seismic coupling in Comsol

I've been looking at the effects of changing the spacer length with Chris. I took Evan's scripts from the SVN and modified them slightly to loop over different spacer lengths. There were some issues with the geometry on the versions of the scripts on the SVN. I made some small changes and was able to replicate Evan's previous results and committed these changes.

I ran the simulation over several values of spacer length (L), support separtion, and support angle. The support seperation was normalized to the spacer length. 0 would indicate that the supports were both placed halfway down the spacer, while 1 would indicate the supports were at opposite ends of the spacer. The support angle is defined as the angle the support makes to the horizontal plane (e.g. 0º would indicate the cavity was supported horizontally, 90º would indicate the cavity was supported vertically). I fixed the outer radius (R) at 1". The mesh size was 4, which is less refined than what Evan was previously using (2).

Attached is the output of these simulations. Each plot is the result of a different support angle. The x and y axes are the normalized support separation and the spacer length respectively. The color indicates the frequecy noise generated by applying 1g of vertical acceleration. Each plot has the same color scaling.

Just as in the previous results, the more horizontal the supports are, the lower the support noise. We also see that the best positions for the supports are either close to the midpoint or near the ends. In general, the shorter cavities have more noise but only by at about a factor of 2 at most. The most significant effect comes from the angle of the supports.

I'm not entirely sure what's happening in the top left corner. It looks like supports are so to each other that there's only one element conecting them. I should be able to resolve those with a finer mesh.

  1274   Tue Jul 21 19:14:52 2015 DmitryDailyProgressCOMSOLMore COMSOL modeling

I took ringdown measurements of 1.8in cantilever at 4.3*10-4 torr:

1.8'' Cantilever Q Measurements
Frequency, Hz Q Mode
320 1000 first order mode
1991 625 second order mode
2087 2000 torsional mode

Also I tried to further tweak the dimensions of the pinwheel in COMSOL in order to make it predict the frequencies closer to the measured ones. While being able to match the frequencies of the first order and the second order modes, the torsional mode was always far off. So I switched to the anisotropic model – (100) wafer, cantilevers along [110] and [-110] axes.

Using this model, COMSOL predicts the following frequencies for the pinwheel of thickness 0.0112in with cantilevers of length 2.44in, 2.2in (*no experimental data for this one), 2.04in, 1.86in an width 0.385in:

Calculated Eigenfrequencies of the Pinwheel
Cantilever\Mode First Mode, Hz Second Mode, Hz Torsional Mode, Hz
2.44'' 161 1009 1449
2.2' 210 1316 1671
2.04'' 255 1602 1853
1.86' 328 2057 2124

For the longest cantilever, the frequencies are very close to the measured ones.  The ratios of total strain energy (E_clamp), stored outside of the pinwheel, to the strain energy stored in the pinwheel (E_pinwheel) are the following (for the longest cantilever):

Elastic Strain Energy Ratio
Freqiency, Hz Epinwheel, (J)* Eclamp, (J)* Ratio
161 3016 3.4 1.1e-3
1009 120264 139 1.2e-3
1449 212434 160 0.8e-3

*Actually should be regarded as a.u.

Seems that the low Q cannot be explained simply by the energy leakage into the clamping (however any kind of friction is not accounted for in the model) and this source of damping also does not explain different Q's for the three modes (??? - the leakage is the biggest for the second mode and the smallest for the torsional one - that is conistent with what we see in the experiment, but the relative change is too small).

Then I checked whether we get measured Q's if, suppose, we have a “very bad” surface. I used the first model (only pinwheel and sapphire washers) and added thin (19µm) layers to each surface of the pinwheel. Assuming the bulk loss factor is equal to the thermoelastic loss, COMSOL predicts the following quality factors (for the longest cantilever):

Quality factor with lossy surface
Frequency, Hz Bulk Loss Surface Loss Q
161 2e-5 9e-4 2230
1009 1e-4 9e-4 2044
1449 8e-6 9e-4 2280

Low Q could be caused by a very lossy surface, but still this does not explain lower Q for the second mode and higher Q for the torsional one (??? same as for the clamping loss - the change in Q is consistent, but too small compared to the measured one)


  1273   Thu Jul 16 21:54:10 2015 MattDailyProgressCOMSOLPotential loss sources and more COMSOL modeling

I put together a quick COMSOL model of the Painter 2 resonator that Zack, Nic, and I were testing earlier. The first mode was calculated to be at 150Hz, which matches our data from 7/14 very closely. See below:

Next Dmitry and I began working on simulating the full pinwheel setup, including the SS clamp, PEEK base, and sapphire washers. We calculated the energy stored in each domain using the solid.ws variable in COMSOL as per Nic's suggestion. This involved integrating the energy density over the volume of each mount component. The results are shown in the table below for the first four modes of the resonator:

Eigenfrequeny (Hz) Si engery (J) PEEK energy (J) SS clamp energy(J) Sapphire energy (J)


2088 0.02 0.14 1.16
228 3197 0.03 0.23 5.01
294 5065 0.06 0.42 8.14
391 8600 0.14 0.82 17.33

COMSOL does not perform any displacement calibration by default when operating in frequency domain, so only the ratio of energy values should be considered, not the values themselves.  The important thing to note here is that all of the other materials store much less energy than the si pinwheel. This means that the loss in the PEEK base would need to be 10^5 or 10^6 times larger than the cantilever loss in order to be comperable to cantilever losses. Same thing goes with the SS and sapphire, which are less lossy to begin with. Does this indicate that loss in the PEEK base isn't a problem? If so, we wouldn't benefit from making the switch to macor.

We also did a rough calculation of the Q of the 2.4in pinwheel resonator. Using the formula for thermoelastic loss in the Glasgow paper, we estimated a Q~50,000. We are clearly not there yet, so there must be other important loss sources present in the sysem. Clamping?

Dmitry did some more involved analytical work in COMSOL modeling thermoelastic loss in the pinwheel system. His results show the estimated Q at different modes if we are only limited to thermoelastic losses:

Eigenfrequency (Hz) Q
184 29,583
228 23,103
294 17,995
392 12,875

As expected, these results are all less than the 50,000 estimate since the calculations take the entire system (mount, clamp, washers, etc.) into account, but they're still much greater than the things we've been measuring. The Q decreases as f increase since in this regime the loss angle phi goes nearly linearly with drive frequency omega. What does this tell us? We probably aren't limited by thermoelastic losses.

Today we also reclamped the pinwheel resonator and started measuring the 2in cantilever. We did some quick ringdown estimates at room temp and 10^-3 torr and didn't see any significant changes in the Q. We'll take a couple more careful measurements at lower pressure tomorrow but I'm skeptical that we'll see any improvements from the other pinwheel ringdowns.

  1272   Thu Jul 16 00:31:15 2015 MattUpdateCOMSOLPinwheel modeling with COMSOL and Q measurements

Today Dmitry and I spent some time simulating the pinwheel cantilever with COMSOL so that I could get my feet wet with the software. We calculated the first six modes of the resonator experimenting with different mesh configurations. f0-f3 are the first order modes for each of the four cantilevers (longest to shortest length), while f4 and f5 are the second order modes for the two longest cantilevers respectively. The results are summarized in the table below:

Pinwheel Eigenfrequencies
Mesh Setting f0 (Hz) f1 (Hz) f2 (Hz) f3 (Hz) f4 (Hz) f5 (Hz)
Normal 216.8 281.8 352.8 498.2 1503.8 1947.6
Finer 184.6 231.0 297.1 396.2 1169.9 1460.8
Extremely Fine (5 layers) 182.5 228.1 293.3 390.7 1143.7 1429.1
Extremely Fine (10 layers) 184.4 228.8 294.8 392.3 1155.6 1433.7

Once we were satisfied with our simulation, we took a PSD and some more ringdown measurements of the longest cantilever on the pinwheel (2.4in). We estimated the Q at several eigenfrequencies, and the results are shown in the table below:

2.4'' Pinwheel Cantilever Q Measurements 300K 10^-2 torr
Mode Frequency (Hz) tau (s) Q notes
160 4.5 2,260 fundamental
1003 0.4 1,260  
1440 0.8 3,600  
2005 - - harmonic?
2443 0.4 3,000  
2880 - - harmonic?

  Since the measured resonances didn't match our previous modeling results, we began tweaking parameters in the COMSOL simulation to better fit our measurements. Changing cantilever dimensions within the machining tolerances had little efffect on the eigenfrequencies. However, changing the sapphire washer size (fixed in our simulation) had a significant effect on the eigenfrequencies. With smaller washers we predicted eigenfrequencies of 166Hz, 1045Hz, 2941Hz, etc. These are closer to the actual frequencies we measured, does this tell us something about our clamping setup? Our sapphire/SS washers may not be completely flush with the pinwheel so we might have a smaller clamping area than designed. Some of these simulations for the first few modes are attached.

After taking these measurements we opened up the chamber and flipped the washers with the hope of getting a better mounting setup the second time. We did a quick Q estimate of the fundamental mode at 10^-4 torr and didn't see any significant improvement.    

  1271   Sun Jul 12 01:21:22 2015 ranaUpdateSiFi - ringdownRingdown measurement with new clamp

Probably right, but hard to be sure.

20% sounds like a lot, but its really not if the tolerances were not held in the machining. We wouldn't be at all surprised if the frequencies were 5% off from spec.

It should be easy to measure / calculate if the beating is an issue. It mostly depends on the coupling which is there through the clamp; hard to quantify I think unless COMSOL does time domain ringdowns.

Also, I wonder why PEEK is necessary at all. Can we not replace that with a hard base? e.g. a hard ceramic or some easily machinable glass? I don't have a great suggestion for material, but I expect that there is one out there (low conductivity, high Young's modulus). Or is the base already so big that the PEEK being soft doesn't matter? In this paper (http://arxiv.org/abs/gr-qc/0504134) from Glasgow, they use Macor.

  1270   Fri Jul 10 11:44:27 2015 ZachUpdateSiFi - ringdownRingdown measurement with new clamp

The modes aren't close enough. We designed these to be different in frequency by increments of 20% for exactly that reason.

Spectrum-wise, we can measure each cantilever individually, but the response of one to another's fundamental mode should be very small...


I wonder about the mode beating effect on the ringdown. Can you post the spectrum so we can see what the overlap is? If the frequencies came out too close, then the Q would look low because of energy sloshing.


  1269   Fri Jul 10 11:12:16 2015 ranaUpdateSiFi - ringdownRingdown measurement with new clamp

I wonder about the mode beating effect on the ringdown. Can you post the spectrum so we can see what the overlap is? If the frequencies came out too close, then the Q would look low because of energy sloshing.

  1268   Thu Jul 9 17:45:03 2015 MattUpdateSiFi - ringdownRingdown measurement with new clamp

Today Zach and I tried out the new clamp setup. The clamp is much thicker than the previous post design and has a lip to constrain the sapphire washer. There's also a large SS washer on top to (hopefully) distribute the force of the 1/4 20 screw more evenly. We forgot to take pictures of the new assembly, so I've attached the SolidWorks model: 

 We estimated a 1/e decay time tau ~5s and a resonant frequency f0 = 150Hz, so the Q was around 2,300 at room temperature. This is comparable to the old cylindrical mount with sapphire washers. no

  1267   Wed Jul 8 20:07:21 2015 ZachSummarySiFi - ringdownSummary of all ringdowns thus far
Resonator f0 [Hz] T [K] Q Comments

Si, 300 um uniform thickness

(symmetric design)

241 300 530

From Marie's original measurements.


" 160 1790
" 77 10,500

Same as above,

but with sapphire washers

250 300 4300

Washer didn't fit around clamp lip.


" ~100 8000

Same as above,

but steel washer added to make contact flush

250 ~90 20,000 CRYO:1197, CRYO:1198

Same as above,

but in large rectangular clamp rather than cylindrical one

250 300 2000

Attempt at removing potential coupling to lossy PEEK


Same silcon resonator as all above,

but sandwiched between 2 other 300-um pieces of silicon

(rather than sapphire washers)

340 300 6800

Thinking non-flatness of sapphire could have been the issue.


Taiwanese Glasgow-style cantilever

92 um thickness

106 300 14,000

Possibly gas damping limited.


" 95 6.9 x 105 CRYO:1213
" 300 1.2 x 105

Improved clamping with spacer, waited for much lower pressure


" 100 6.7 x 105

No significant change at low temp vs. pre-spacer (2 rows up)


" 90-120 0.5 - 1 x 106

Measured continuously over temperature range


" 120 4 x 105

In new rectangular clamp. No improvement (slightly worse, in fact).


Painter prototype 1 ~350(?) 300 4000

In old rectangular clamp


Painter prototype 2 ~150 300 40,000

In old rectangular clamp


" 100 60,000 CRYO:1265
" 100 30,000

After HF surface treatment (worse!)


Flower Mound propeller resonator

300 um, uniform thickness, asymmetric

145 300 2000

In old cylindrical mount, with newer sapphire washers



  1266   Thu Jul 2 16:15:40 2015 ZachDailyProgressSiFi - ringdownQuick update on recent measurements

[Matthew, Nic, Zach]

Over the last few days, we've done 2 significant measurements:

Flower Mound propeller resonator sandwiched between sapphire washers at room temperature

We clamped one of the propeller resonators between two of the new, larger sapphire washers, using the cylindrical mount in the room-temperature chamber. Unfortunately, the inner diameter of these washers is also too small to fit around the lip in the bottom part of the clamp (as it was with the old washers---see CRYO:1197), so I needed to reuse the steel washer that I fashioned before.

We measured the Q to be a rather low ~2,000.

It's unclear exactly why it should be so bad, but we have never measured a good Q with these thick, uniform resonators. It may be caused by the effect I noted in CRYO:1198, where the mode is highly coupled to the relatively soft and high-loss PEEK. It could also be from clamping non-idealities from the fact that the bottom sapphire washer does not sit flush on the clamp over its whole surface.

Matthew has designed a revised version of the cylindrical clamp. The main improvements are 1) it is bigger and therefore should not allow for as much coupling of the silicon modes to the PEEK base, and 2) it will be made to accomodate the sapphire washers we have. Nic put the order in and it should be in late next week.


Painter fab prototype 2, after surface treatment, in new polished block clamp at ~100 K

I finally got a chance to meet with Justin again to do a surface treatment of the second prototype resonator (the one in CRYO:1264). This consisted of 10 minutes in "piranha" etch (this is 3:1 hot H2SO4 and H2O2), followed by 10 minutes in room temperature HF. According to Justin, this is a passivation treatment that only affects the surface chemistry of the resonator; it does not repair the surface in terms of the roughness. Since our leading hypothesis was that our low Q for this thin resonator is related to the rough surface, I didn't have high hopes for a big improvement.

Matthew and I put it in the cryostat yesterday and cooled it overnight. This morning, we measured a Q of around 30,000 at 100 K. This is about a factor of 2 lower than before, which is certainly surprising...

Matthew did some real data analysis on this ringdown, and I'll have him attach it to this post.

Update (Matt):

I just attached a couple data analysis plots that I used for determining Q. I calculated a Q of ~28,000 at f0 = 145Hz. I'm taking an FFT of the ringdown measurement, filtering around the resonant frequency (in the time domian), and then fitting an exponential to the filered signal to measure tau. 



  1265   Wed Jun 10 18:09:39 2015 ZachUpdateSiFi - ringdownSecond prototype cantilever cryogenic Q not great

I measured the Q of the second prototype with the new clamp in the cryostat at low temperature. The result was ~60,000, which is not much better than what was measured at room temperature (see quote from last post).

My next hypothesis as to what's going on is that the new cantilevers are thus far dominated by surface loss. If you compare the apparent quality of the surface of the Taiwan cantilever (photo in CRYO:1211) with those of the first (CRYO:1260) and second (CRYO:1264) prototypes, it's obvious that it's much better, at least visually. There is also some clear macroscopic blemishing on the first prototype, which could be leading to its abysmal Q of 4,000 at room temperature.

Two parallel courses of action that are underway:

  1. Use an estimate for the roughness of the etched portion of the prototype cantilevers to see how bad the surface loss would have to be to explain our measured Qs. I'm waiting for Justin's answer on this.
  2. Start trying some surface treatments. It could be that we just need to get a little better at fabrication at this point, but it can't hurt to try a treatment and see if we get a noticeable chage in Q. This is certainly only worth it for the second prototype, which at least appears fairly uniform in roughness.

We are also going to get started with a third prototype soon. For that, we'll be working on our cleanliness procedures.


Prototype 2 in old clamp at room temperature: Q ~ 40,000. Pretty good. The best we've seen is around 100,000 with the Taiwan cantilever (see CRYO:1216), but this was after pumping much longer. The Taiwan cantilever showed Q ~ 40,000 at similar pressures, so it could be that this prototype is as good.


  1264   Sun Jun 7 17:33:31 2015 ZachUpdateSiFi - ringdownNew clamp received, second prototype cantilever made, some measurements

This is a recap of some things that have happened over the last couple days that I have yet to elog.

New clamp

We received the new clamp for the ringdown cryostat. The drawings are attached, but to jog the memory:

  • It has similar dimensions to the old clamp
  • Mounts to the same PEEK insulating base as the old clamp
  • The clamping surfaces are polished to mirror finish
  • There are tapped holes near the corners for guide rods to ensure level clamping. The guide rods themselves were ordered from McMaster-Carr, but they were lost by CIT mail services. I'll find them this week.
  • There are 3 pairs of 8-32 holes for clamping (one pair at any given time---just to be able to clamp in different places).
  • Holes on the side for mounting of the power resistor for heating
  • a channel on the side for taping on the RTD

It looks pretty good. Here is a shot of it, and then a second shot focusing on my camera in reflection, to give you an idea of the finish quality:



New prototype cantilever

Justin and I made a second prototype cantilever. It is 50 mm long, with 10 mm of clamping/mirror region on either side. We estimate the thickness of the central region after etching to be ~200 um.

As you can see, the "un-etched" regions were etched a bit, particularly on one face of the cantilever. We believe this to be caused by cleanliness issues that are difficult to avoid when prepping/etching both faces in one go. We're going to iterate on this somewhat, but we may need to do the prep and maybe the etch on one face at a time.


I also did some very rough testing of the optical quality of the tips. Shining the 1550nm laser on it, I got some strong etalon action. The max transmission I got was around 90%, and the minimum was on the order of a few percent---I need to do this more methodically. The TRANS and REFL beams seemed to have good quality, but I'd like to see them on the CCD to look for aberrations.

Some ringdowns

With these new toys, I did a few more ringdowns.

Prototype 1 in old clamp at room temperature: Q ~ 4000. Not so great.

Prototype 2 in old clamp at room temperature: Q ~ 40,000. Pretty good. The best we've seen is around 100,000 with the Taiwan cantilever (see CRYO:1216), but this was after pumping much longer. The Taiwan cantilever showed Q ~ 40,000 at similar pressures, so it could be that this prototype is as good.

Taiwan cantilever in new clamp at 120 K: Q ~ 400,000. This is about what we measured with the old clamp (see CRYO:1230). The point of this measurement was to see if the result improved with the new clamp, with the hypothesis that we were limited by clamping effects from the poor finish of the old clamp before. At this point, this seems not to be the case.

Thought sandwich bread time.


  1263   Sun Jun 7 17:06:49 2015 ZachLab InfrastructureCryoLN2 refilled, new dewar #200

(on Thursday...)

  1262   Wed Jun 3 19:39:18 2015 ZachNotesSiFiSi structure fabrication process

I went into the lab with Justin today to make our second prototype resonator. I am writing the procedure down here from memory. I will correct/update this as we continue.

Note: of all the steps below, only the photolithography is done at KNI. The rest are done within the Painter labs.

Initial prep

  1. Peel specimen from the adhesive packaging it is attached to by the dicing company.
  2. Rinse with acetone and isopropanol.

Photoresist "spin and bake" (must be done twice: once per side)

  1. Pre-treatment.
    1. Soak specimen in weak solution of H2O2 and NH4OH for around a minute.
    2. Dry.
  2. Primer.
    1. Attach specimen to spinner (with suction).
    2. Apply primer solution with disposable pipette using designated pump. Apply evenly over surface, enough so that the meniscus nearly reaches the edges.
    3. Spin at 3000 rpm for 1 minute.
    4. Bake at 110 ºC for 1 minute. (All bake cycles are in air on a hot plate.)
    5. Bake at 220 ºC for 5 minutes.
    6. Allow to cool before proceeding to photoresist.
  3. Photoresist
    1. Attach specimen to spinner (with suction).
    2. Apply photoresist solution with disposable pipette using designated pump. Apply evenly over surface, enough so that the meniscus nearly reaches the edges.
    3. Spin at 3000 rpm for 1 minute.
    4. Bake at 110 ºC for 2 minutes.
    5. Allow to cool before proceeding to photolithography.
  4. Until controlled exposure (photolithography) and development, protect the sample from light by wrapping the holder in foil.

Photolithography (this must also be done once for each side)

  1. Mask preparation (only needed when using transparencies, as we are currently)
    1. Place transparency with mask pattern on square glass blank. Center as best as possible to ensure that you'll be within the alignment actuator ranges later.
    2. Slice a piece of scotch tape into thin sections.
    3. Tape the transparency to the glass blank in all four corners, with the transparency as flat against the blank as possible (to avoid diffractive aberrations).
  2. Mounting the mask
    1. Turn on external suction pump.
    2. Install blank with transparency attached to it onto the sliding palette. Transparency should be face up, since the palette will be flipped upon installation into the instrument.
    3. Engage suction.
    4. Flip palette and slide into the instrument.
  3. Mounting the specimen
    1. Use 2" wafer holder.
    2. Install "chip holder" adapter. This blocks most suction holes that would be used for a full 2" wafer, leaving only a few near the center to hold a smaller chip. (This works for our 50-mm-long cantilevers, but we will have to use a different method to hold anything longer.)
    3. Place, align, and center specimen as best as possible onto the holder, then slide into the instrument. Through the window, one should now see the inverted mask held in place above the specimen.
  4. Alignment and exposure
    1. Set exposure type. Some high-resolution lithography requires the mask to be in direct contact with the specimen; we don't need such resolution, so we choose keep a ~50-um gap between the mask and the specimen.
    2. The instrument now fully engages the specimen against the mask (vertically, from below), then backs off the appropriate amount.
    3. Choose an appropriate microscope objective and locate the mask and/or specimen edge on the viewer screen.
    4. Find a good edge for alignment, then align the specimen (movable) to the mask (fixed) in both angle and position with the knobs. Scan the viewer around to ensure good alignment.
    5. When satisfied with alignment, engage exposure. We used 40 seconds at 365 nm, but this depends on the photoresist. The viewer will retract and the illuminator will engage for the designated time, then it retracts and the viewer reengages.
  5. Clean up after yourself and leave instrument as you found it.


  1. Dip the specimen in the developing solution and swash for 10 seconds.
  2. Remove and blow-dry with N2 gun rapidly.
  3. Repeat (1) and (2) until desired pattern is very clear (this is the first point in the process where the pattern should be visible to the eye). The un-exposed regions should be a clean Si surface, while the exposed regions should appear slightly darker.

Etch (to be continued...)

  1261   Tue Jun 2 16:38:17 2015 ZachUpdateSiFiSuspension blades acquired

Norna very graciously acquired me some spare aLIGO prototype blade springs. I now have 5x D080761 HSTS lower blades, which is enough for the experiment and one spare.

These are meant to hold the HSTS payload of around 6 kg, so that makes the discrepancy I mentioned in CRYO:1258 a little better (the OMC blades are formed to support 7 kg, so we have less mass to make up for now). The next decision will be whether to make our 7.5" breadboard out of aluminum or steel.

  1260   Wed May 27 19:23:25 2015 ZachUpdateSiFiFirst prototype cantilever made

Justin was working this week on the first barbell-type cantilever, and he handed me the finished product this afternoon. It's a bit rough around the edges (literally), but I think it's a good start.

To make this, we had to do the photoresist masking step using a pretty low-budget scheme of high-resolution printing on a standard transparency. Justin thinks the print shop we went to didn't do a great job, and as a result we have some areas on the resonator that have been over- and under-etched, leaving some burrs and pits.

Below are some photos. The first two show the surface of the cantilever with different lighting, so you can get an idea of what it looks like. The last one is a shot focusing on the illuminating lamp in reflection of the cantilever surface---as you can see, the end sections (which are ideally not etched at all due to masking) largely retain their optical polish, while the thinned center region becomes a bit rougher, as expected.

Tomorrow, we're going down to a different print shop that Justin has had positive experiences with in the past. Eventually, we will make a high-quality mask out of glass, but it's worth using cheap transparencies for the first few rounds.

  1259   Mon May 18 12:34:06 2015 ZachSummarySiFiGWADW 2015 poster

The GWADW poster can be found at G1500651. Also, here is the rendering I made for it:

  1258   Mon May 18 12:31:25 2015 ZachPhotosSiFiA taste of things to come...

EDIT (ZK): The rough estimate for the payload as it is currently envisioned is about 3 kg, considerably lower than the OMC's 7 kg. This can be brought up closer to 5 kg by making the breadboard out of SS instead of Al. That makes it a little closer to a reasonable pre-trim weight, but it's also the case that the clamp-and-cantilever side and the macromirror side are balanced to below 10%, as it turns out, so this would just be a 2-kg dead CM addition. Back to the materials question, the bending mode issue is a wash, since the frequencies are roughly the same in both cases here.

So, I'm going to talk to Gabriele and see what the best option looks like with this information in mind.

EDIT 2 (ZK): ICS shows 27 clean spares, in addition to the 12 in aLIGO assemblies and one that apparently did not go through C&B. Need to see exactly how Gabriele went about getting spares. I also used the blade calculator spreadsheet to find that a mass of 5 kg would leave a residual (upward) deflection of around 1.6 cm. This is well within the headroom and could help to maximize the pendulum length.

True. Without having done a detailed calculation, it seems to me that out payload will be of similar mass (maybe slightly higher) to that of the OMC. So---again pending a real calculation---I'm thinking spare OMC blades would be ideal if they are available. But I'll get on the real calculation pronto so that we can get something custom going soon if necessary.


Sounds wise. But if you have a good idea of what's needed in terms of mass for the platform (with appropriate extra trim mass added for margin), then you can get some going into the machine shop queu now and not have to wait for 3 months later on.


  1257   Sun May 17 18:39:48 2015 Rana(?)PhotosSiFiA taste of things to come...

Sounds wise. But if you have a good idea of what's needed in terms of mass for the platform (with appropriate extra trim mass added for margin), then you can get some going into the machine shop queu now and not have to wait for 3 months later on.

  1256   Sat May 16 22:06:55 2015 ZachPhotosSiFiA taste of things to come...

Talking with him, I got the impression that going with spares, if they are available, is the best course. We can probably do better by designing new ones, but I think this can be a time-expensive process. I'm going to see what options we have "off the shelf" and then we can consider a custom job.


Might be able to get blades made instead of just using spares. Crackle might have some spares and they also have a place they get them made. Gabriele ought to be able to give us advice on whether to use as is or modify by adding constrained layer damping ala the ISI.


  1255   Sat May 16 15:34:47 2015 ranaPhotosSiFiA taste of things to come...

Might be able to get blades made instead of just using spares. Crackle might have some spares and they also have a place they get them made. Gabriele ought to be able to give us advice on whether to use as is or modify by adding constrained layer damping ala the ISI.

  1254   Wed May 13 06:37:30 2015 ZachPhotosSiFiA taste of things to come...

I've been working on an early-stage mockup of the in-vacuum setup for the main experiment for my GWADW poster. Here is a cutaway shot of what I have so far and also a detail of the suspension. The breadboard at the bottom is what I'd like to get made custom by ThorLabs (plus a couple mods we'll have to do ourselves, like the holes for the OMC-style captive ferrule joints). The blades are OMC bottom-stage blades, but HSTS lower stage will also work. We need to see what spares are available.

More later.


  1253   Thu May 7 03:35:49 2015 ZachElectronicsSensorsRL readout PD with humbucking

I got the second inductor in recently, so I tested out the humbucking circuit. Here are the results of a quick trial:


  • As you can see, having the humbucking coil greatly reduces the influence of noise pickup---particularly the flurry below a few hundred Hz.
  • As measured by the mains lines, the rejection is not that great at the moment, at something like a factor of 10. This measurement was made simply by putting the reversed coils next to one another, but it's likely that I'll have to do some more tuning to get them matched to the ~1% level I'd like.
  • Aside from residual pickup, the noise is exactly what is predicted by LISO
    • A caveat is the noise peak feature at ~40 kHz, which is probably something to do with the op amp loop. Investigating...
  • Considering that this is in a relatively high-EM environment, things are already looking pretty good. The DC Z of this circuit is ~100 ohms, so it can measure RINs of ~10-11/rtHz above 100 Hz yes 

Hartmut says that they see excess noise in their circuit when using it for DC readout, which he believes is likely Barkhausen noise. So, the next thing to do is inject a reasonable, mildly fluctuating DC current and investigate the noise performance in that case. The only method I know for creating a current with low enough audio noise to see this level is using the buffered active-lowpass-reference scheme (e.g., what we used to drive the LEDs in our OSEM testing days and what is used as the bias on the M2 board). Not thinking too much in advance, I connected the M2 bias (5 V) to the circuit, which therefore drove a ~50 mA current through the coil, and the output noise was not changed significantly. In hindsight, this is not surprising: the mentioned current generation scheme is actually just a voltage reference, and so the voltage atop the inductor was actively fixed at 5 V, so that what I saw at the output of the entire circuit was just the noise of the 5 V reference circuit.

So, I need a true current source, and I'm not sure a good enough one exists to inject tens of mA without swamping the dark noise of this detector completely. I think the best I can do is to use whatever I can and then just verify that no excess noise is observed above the input current's own noise. Perhaps it's easiest to just do it optically: using balanced PDs, stabilize the light to around the shot noise level with a normal M2 transimpedance channel, then read out the OOL PD with the RL humbucker.


So, we are left with the problem of being highly sensitive to a signal injected into the coil, but not to one induced by external fields. Luckily, the guitar industy has had a solution for almost exactly the same issue for about 80 years now: the humbucker.


I've ordered another identical inductor to see what we can do.


  1252   Wed May 6 18:01:02 2015 nicolasLab InfrastructureDrawingsCryostat base plate

To lift the cap screws of the cryostat off the table, I've designed this base plate.

It has 8 holes that are coincident with the screw holes on the bottom of the cryostat. 4 of them let the full cap pass through, the other 4 will fasten the cryostat to the base plate.

There are slotted holes for mounting to the table, but dogs could also be used.

The four corners are slightly thicker than the rest of the plate (0.5mm) to make feet.

Edrawings and solidworks files are in the SVN @ 40mSVN/SiCryoSus/Designs/SiFiDewar/BasePlate

  1251   Tue May 5 17:10:14 2015 ranaUpdateSiFiwafers received

NOVA wafers are here and on the VIP desk in Koji's office.

ZK: Wafers unpacked and stored in 050.

  1250   Thu Apr 30 18:00:47 2015 ZachUpdateSiFiDiced wafers received

We re-received the 6" wafers we had sent to American Precision Dicing to be chopped up last week. One wafer was cut into 10 cm x 1 cm rectangles, and the other into 5 cm x 1 cm ones. They came back in some super cool packaging.


  1249   Thu Apr 30 17:53:20 2015 ZachCryostatPurchasesGetters gotten

Being the go-getters we are, we got our spare getters gotten. They'll be stored in the cabinet for when we need them.


  1248   Fri Apr 24 04:31:48 2015 ZachLaserSiFiCavities locked simultaneously, transmission setup built

I was running into some issues optimizing the servos with the present actuation setup, and I had a conversation with Dmass about another short-term solution (more on that at the end), so I figured I'd try a simpler approach to getting both beams at the output for the time being: I knew I could get one cavity locked with a simple SR560 servo, so I decided to try locking the other in the same way. I didn't expect this to work very well, since the other diode driver I had to work with only has ~1-kHz bandwidth, but miraculously it did. The lock is a bit shaky, but it holds indefinitely and relocks readily as with the other cavity. I guess there's just enough less noise with the cavities inside the cryostat.

With the cavities locked, I went about building the transmission setup. The beams are both directed west via 90R/10T mirrors, where they then each pass through a QWP to return them to linear polarization and then a HWP to set them to S, which is the favorable polarization for optimal 50/50 splitting with the beamsplitter we are using. The waveplates were optimized using an analyzer PBS, and I got the polarization contrast down to about 0.3% in both cases. I used geometry to make sure the beams were combined at the same gouy phase. One ouptut of the beam-combining BS is then steered and focused onto a 1611 PD, and the other output will be put on an auxiliary PD for DC TRANS diagnostics. I also decided to overload the CCD camera as a dual-beam sensor by just pointing both of the leakage beams through the first steering mirrors at the camera, side by side (as seen in the photo above). The beams are positioned relative to each other on the screen as the cavities are seen from the rack (E beam on the left, W beam on the right).

The whole setup looks a bit like Stiltsville right now (explained here for non-Floridians), which isn't the long-term plan. When I get a feel for how much space I really want, I'll get a breadboard and just elevate it by the ~2" required. An added benefit to that scheme is that the whole thing can be removed in one piece for access to the cryostat.

Here's a glamour shot of the I/O optics:

I aligned the combined beams and tried to get my first transmission beat, but didn't have any luck right away and needed to call it quits. The procedure I had planned for doing this in the future is:

  1. Use the frontal beat to bring the two beams close together in frequency. One easy place to do so is at the RIO test report temperatures.
  2. Scan one beam until a resonance is found and then lock.
  3. If the frontal beat has been lost (fairly likely, as the frontal beat PD is an 1811 with BW ~ 125 MHz, while the cavities have an FSR of >2 GHz), scan the second laser until the beat is found.
  4. Once beating, scan the second (still unlocked) laser in both directions and find the closest resonance, then lock.
  5. In the worst case, the two cavities' closest modes are ~1 GHz apart (i.e., FSR/2), so it should be possible to see a beat if the procedure is done properly. For the real experiment, we'll have some slow length actuation (e.g., via temperature) so that we can bring the two cavity modes as close together as we want.

Returning to the actuation situation, Dmass actually prefers that I hijack his diode drivers temporarily, rather than trade for his ITC510 unit (even though he is not using the diode driver side of that unit, he is worried about having to retune the temperature loops with a new unit). This is actually way better for me, as well, since it will really let me tune up my servos the way I'll want them when I grow up and get my own Rich drivers. Now that the optical layout is more or less complete (at least for the short term), I can focus almost exclusively on the loops. There are some spare long BNC->DSub cables for exactly this purpose that I can use to drive my lasers without moving the drivers from the CryoCav rack, so I'll get on that shortly.


  1247   Thu Apr 23 03:09:49 2015 ZachDailyProgressSiFiWindows installed, cryostat top with cavities placed in body, cavities realigned, locked

These things happened over the last few days but haven't been elogged.

Windows installed

I installed the AccuGlass windows into the flanges, but ran into some trouble as one of them cracked. It happened on the 4th one, and I did not use any more force than seemed reasonable given that the glass-on-o-ring contact has to hold the vacuum. I think we might need to get some teflon gaskets to soften the metal-glass contact.


Cavities installed into chamber body, realigned, locked

Despite the above, since we had 5 windows, I finished putting them on the cryostat and went about installing the cavities (mounted to the coldplate and cryostat top) into the main body. I transferred the top from the open-air test rig (see CRYO:1244) into the chamber body, then positioned the whole assembly according to some reference marks I made on the table. Using a fast lens in transmission of one of the cavities, I was able to align the cryostat well enough by hand to see some high-order flashing (the test rig is ever-so-slightly taller than the assembled chamber, so what I saw was mostly TEM0n modes). With some tweaking of the input alignment, I aligned each beam to its cavity's TEM00. Here is one cavity locked while in the chamber:

It was all very deterministic, so that inspires some confidence in the procedure.

Other stuff

I also tuned up the W RFPD to match the temporary testing TF I set on the E one (readout 30 MHz, notch 60 MHz---see CRYO:1242), and used it to lock one of the cavities.

I attempted to modify the PDH2 from gyro configuration into a useful one for testing here, but I may have changed too many variables at once; I knew I wanted some more gain in the ~1 kHz region, so I used an extra P/Z than what is available with the uPDH box to do so. Somehow, I'm not getting any good locking action even though the TFs are identical at DC and near the UGF of the known working loop at ~30 kHz.

Also, I remembered that there's no hope to do any good locking with the other (standalone, LDC201C) laser driver that I have, since its bandwidth is about a kHz (CRYO:1205). I'm asking Dmass if it's OK if I trade him my standalone temperature controller (TED200C) for one of the integrated units (like the one I have one of), since he's only using the temperature controller half.

  1246   Tue Apr 21 21:30:50 2015 ranaLab InfrastructureDrawingsLab Layout circa Feb. 2015 (according to Nic)

Heavy Duty (14 gauge steel) cabinet w/o glass doors, but with door storage ordered.

From CryoLAB: April 21, 2015
  1245   Tue Apr 21 14:19:16 2015 ZachLaserPurchasesFiber isolators arrived

We bought some IO-H-1550APC fiber isolators from ThorLabs last week. They arrived today:

  1244   Fri Apr 17 05:15:26 2015 ZachLaserSiFiTest cavities rebuilt on cold plate, aligned, locked (separately)

Given the level of noise I saw in air with the table-mounted test cavity yesterday (CRYO:1242), I decided to move straight into in-vacuum testing.

The first thing I needed to do was set up an in-air support system for the top half of the cryostat for mounting and alignment (i.e., I needed to suspend the cold plate in the position it will be in with the cryostat closed, but with access to the work area). By what I can only describe as a magical coincidence, the height of the upper/lower joint on the cryostat is almost exactly the height of two long 1" pedestals stacked together, and there are 8-32 taps for the cryostat joining that can be used to mount them (huzzah!):

I temporarily stored the getter in a ziploc bag for the duration of this preliminary testing, then positioned the lifted coldplate over the test cavity area. Then, I went about building the cavity in sleeping-bat position on the cold plate. The cryostat windows are 149 mm above the table (145 mm from the bottom of the cryostat, but lifted another 4 mm by the caps of the screws holding the floor on, which are not countersunk). Before building the cavities, I assembled perisocpes to lift the beams from 4" to 149 mm using discrete components.


I was a bit too optimistic about space when building the test cavities on the table, so I needed to shorten them a bit to around 2.5". Because of this, and because the distance to the waists changed, I needed to recalculate the MMTs. The solution only shifted the lenses by a few inches:

Other solution:

mismatch: 0.00015463
w0x = 200.5644 um 
w0y = 200.5644 um 

lens 1: f = 103.2118 mm
lens 2: f = 103.2118 mm
d1 = 15.2397 cm
d2 = 25.9767 cm
d3 = 74.0136 cm
(Total distance = 115.23 cm)

After realigning the E cavity, it locked again with no trouble using the same servo and settings (note: this shot was actually taken before building the W cavity and periscope):

The lock was actually quite a bit stabler seeming than yesterday's, perhaps due to some mechanical low-passing from the support system. I'm hoping this gets better, not worse, with the full cryostat in place.

After building the W cavity and periscope and aligning, I borrowed all optoelectronics from the E path (criss-crossed laser fibers, swapped the E RFPD in, and used the same electronics chain) and it, too, locked with little trouble, though it seems like the modematching is somehow not as good on that path yet. Here are preliminary control spectra from each cavity (note that these were taken at different times, which I'm hoping accounts for some of the slight differences in noise features):

The control spectra are plotted alongside the michelson displacement noise measurement from the other day (CRYO:1238) as well as the PMC-derived laser frequency noise measurement from February (CRYO:1205). Some things to note:

  • Admittedly, the W cavity measurement was taken hastily at the end of the night. It looks like the UGF was rather low and some gain peaking accompanied it at a few kHz.
  • By and large, the displacement noise looks better with these cryostat-mounted cavities than with the table-mounted michelson. In particular, there seems to be some filtration above a few hundred Hz, which I think greatly improved the locking stability today vs. yesterday on the table.
  • The E cavity spectrum, which is more trustworthy and which should have been made with a UGF near 100 kHz, intercepts the expected laser frequency noise level above a kHz or so.
  • Note that the red trace sits on the dark noise level above 4 kHz, and its being near the laser frequency noise level on this plot is pure coincidence (the michelson wouldn't have been sensitive to this).

All in all, this looks pretty reasonable and things are promising for the in-vacuum testing. yeslaugh

Here is a parting photo of the new layout:

For tomorrow:

  • Install cryostat windows
  • Close cryostat and pump
  • Tune W RFPD and PDH2
  • In-vacuum simultaneous lock
  • Build TRANS setup with beat
  1243   Fri Apr 17 04:39:25 2015 ZachUpdateSiFiTwo 150mm wafers sent for dicing

I sent two of our 150mm (~6") wafers to be diced by American Precision Dicing (Justin's recommendation). One wafer will be cut into 100mm x 10mm rectangles, and the other into 50mm x 10mm rectangles.

Justin gave me a nice wafer holder he didn't need (left), so I've stored all but the two I shipped in that one, and shipped the two wafers using the crappier one they came in from University Wafer (right):

While I had the holders open, I took this shot of the reflection from one of the wafers to demonstrate the niceness of the polish:

  1242   Thu Apr 16 15:58:02 2015 ZachLaserSiFiGyro RFPDs installed, tuned, E cavity PDH tuned and cavity locked

Well, I got partway through the plan for the day, anyway.

Rather than align the W cavity as I had done the E, I decided to continue working on locking the E cavity so that I could copy all the work wholesale to the other side at the end. The first thing I did was to install the gyro RFPDs. To facilitate this, I installed a NIM crate in the bottom of our rack (one of the spares that was stored under the gyro table). This will be used to power the PDs, the PDH2 board, and some of my homebrew filters/preamps if necessary. After powering what became the E REFL PD up, I installed it and focused the REFL beam onto it.


I checked to see if either of our crystal oscillator frequencies were within the tuning range of the PD as it's stuffed now. They weren't, so I just tuned it to 30 MHz (and notched 60 MHz) for temporary testing. Recall that these are aLIGO-style PDs that don't have maximal gain at the resonant readout frequency. In practice, they are tuned by looking at the notch seen by the anode, rather than looking at the readout node as seen below. (Note: the delay seen is consistent with the optical path lengths through the fiber and free space).

I then did some playing around with an SR560 servo, moving the pole and gain until I got some weak locking action. Then, I systematically searched the parameter space until I found the best stability (still bad, though). Using that information, I built into the uPDH box (#1437) a TF that had similar gain in the UGF target region of ~50-100 kHz, but much more low-frequency gain. This ended up being something like zpk([10k,10k], [50, 50], 1000), where the cavity pole at ~40 kHz returns the loop to 1/f above there. I would have put one of the servo zeros at this frequency, but I had too large a capacitor for it to make sense due to the smallness of the resistor needed---this is something I can change if necessary.

Plugging that bad boy in and playing with some attenuation before and after the servo, I got a reasonably stable lock, but nothing too stellar. It holds for many minutes, but it is in a delicate balance (often not very balanced) between lots of unsuppressed audio noise and some instability in the ~100-kHz+ band due to the plant. This is exacerbated by the low bandwidth of the driver (see CRYO:1205---I'm using the ITC502 at the moment). I tried doing some feedback using the bias tee input, but I wasn't able to lock at all with this method. Maybe I need to do some high-frequency crossover to it while keeping the low-frequency actuation going through the driver. Here is a shot showing the transmitted beam on the camera and card, as well as a scope trace of a lock acquisition:


In the above, GREEN is TRANS, CYAN is error, and MAGENTA is actuation. The error signal is dominated by periodic oscillations at >100 kHz, while the transmission shows plenty of audio-band noise.

Given the measurement that I made the other day on essentially the same physical system (CRYO:1238), it looks like the audio-band RMS is on the order of almost a nanometer, which is ~50x the linewidth of this cavity. So, I need more gain or less noise at 1 kHz. There are two options:

  1. Switch to the PDH2 and add rapid rollup below ~30 kHz with up to 4 P/Zs
  2. Just go to vacuum already, since we don't really need to contend with this much noise in the end

Since I want both sides to be roughly balanced (i.e., same loops, etc.), I'm tempted to just do (2). I've already got everything pretty well aligned and I know the electronics TFs are not crazy, so I think that's what I'll do.


For tomorrow:

  • Repeat the last step above for the W cavity
  • Install and tune the gyro RFPDs
  • Determine appropriate PDH servo TFs and modify boards
  • Lock cavities
  • Build transmission beat setup


  1241   Wed Apr 15 01:19:14 2015 ZachLaserSiFiInput paths, frontal beat, test cavities set up and aligned, ready to lock

I bought and received the last optics I was missing (90R/10T BSs) last week, so I began building the real experiment today. Here is what it looks like so far:

Frontal beat

The first thing I did was build the frontal beat, which we will use to locate and adjust the beams' frequency offset without having the cavities locked. This is fit in a rather compact layout between the main beams, using a 10% pickoff from the second mirror in each path. The beat alignment DOFs are matched by the actuation offered by 1) the extra steering mirror in the E (left above) beam path and 2) the combining beamsplitter. A further steering mirror puts the beam on the PD after a focusing lens, and the (small) reflection is dumped. Here are the beams beating near 50 MHz:

Test cavities

We want to use some dummy cavities to set up and test our electronics. To make these, we've used some of Dmass's spare fused silica ATF-coated optics (coating run V6-593/594---scans attached). I chose to use two 50-cm mirrors for each cavity, with a length of ~4" (this is just roughly as big as you can make it with two big mirrors within the vacuum area---note the cryostat boundaries drawn onto the table). The transmission of these mirrors is 0.016%, giving a finesse of ~20k. Here are the arbcav plots:


I temporarily relocated the cryostat to the central table so I could put the dummy cavities within its boundaries on the main table. Installing only the cavity end mirrors first, I finished the main beam paths, aligning the beams along the holes and level at 4" for the main stretch, then installing the steering mirror zigzags to bring the beams in to the cavity longitudes. With the beams centered on the cavity output mirrors, I then calculated a modematching solution. Modematchr came up with plenty of options, but I chose this one because it used only f=100mm lenses (which I had bought specifically for mode matching) and left plenty of room near the cavities for the circulating optics:

Other solution:

mismatch: 0.0002397
w0x = 222.7609 um 
w0y = 222.7609 um 

lens 1: f = 103.2118 mm
lens 2: f = 103.2118 mm
d1 = 11.8781 cm
d2 = 26.4822 cm
d3 = 72.1297 cm
(Total distance = 110.49 cm)

As you can see, the predicted best mismatch is 0.02%.

I installed these (on slotted bases, so they can be adjusted), then verified that the output beams were roughly as expected---they were. I then used the cavity output mirrors to retro-reflect the beams, which served as a further modematching sanity check (since the retroreflected beams agreed transverse-spatially). Next, I mounted and installed the circulating optics (PBSs and QWPs), tuning the initial HWP and the QWP to maximize forward transmission and backward rejection, respectively. Finally, I focused the E REFL beam on a PDA255 for temporary testing.

Now the input beams and end mirrors were aligned, so all I had to do was install and align the cavity input mirrors. Before doing so, I borrowed the 1550nm-sensitive CCD camera from Dmass's setup and placed it behind the E cavity end mirror. With all the lights off, I could very faintly see the transmitted beam, and I centered the camera onto it. The REFL PD gave me a good reference, so just installed the input mirror and directed the prompt reflection back to the PD. Immediately upon doing this, I saw strong transmission flashes on the camera.

Tweaking the cavity and input beam alignment somewhat while scanning the laser frequency, I did a rough TEM00 maximization. The REFL dips indicate ~80% coupling, which I think is as good as I'll bother going before locking. yeslaugh


For tomorrow:

  • Repeat the last step above for the W cavity
  • Install and tune the gyro RFPDs
  • Determine appropriate PDH servo TFs and modify boards
  • Lock cavities
  • Build transmission beat setup
  1240   Tue Apr 14 22:27:18 2015 ranaLab InfrastructureDrawingsLab Layout circa Feb. 2015 (according to Nic)

Some cabinet options (for storage of heav-ish things):

  1. SC-4824 (48" W, 24" deep, 78" H) from TecLab
  2. L47HB from Hanson Lab (same as above, but with glass doors and nicer hinges and rubber door stoppers)
  3. 73324 from hemcocorp.com, same but only 18" deep.
  4. A few on Amazon for ~1200$: http://amzn.com/w/1J0TVPU10HHSN
  1239   Mon Apr 13 11:24:32 2015 ranaNoise HuntingMechanicsPost and base vs. pedestal and fork

Its a nice result. As you say, we are lacking in good writeups about this topic. Mostly they're plots in the iLIGO elogs which were never collated and so are lost to the mists of time...

But I think this is missing one of the major points: the objection to the pedestal/fork combo has to do with the care required to assemble it. When assembled carefully, as you have done, it works well. But taking an ensemble of them in a larger lab has shown from observation, that some fraction of them are assembled poorly. Usually its the last step of nudging the fork which is not done or the final tightening pulls the fork back by a mm or so.

And I think all of these kinds of measurements are including a significant bias. To make a good comparison, we'd have to check that the torque used on the 1/4-20 screws is "good". Also, the finding from assembling the ISC tables back in the late 90's was that the torque used on the optic set screw is important. Too tight makes some distortion of the glass, too loose and you get some springiness. Another fork parameter which I believe is important is the distance between the pedestal and the screw. This removes some of the parameter spacei in the 'arbitrary' positioning capability of the pedestal/fork. When Mike Landry made these measurements ~2000, I believe he found that the springiness of the fork screw was a parameter in the extreme position cases.

Also, in the several previous incarnations of this experiment, we used a little computer speaker as a white noise source to drive the mount under test to make sure that the source noise was not changing.

  1238   Sat Apr 11 23:42:03 2015 ZachNotesMechanicsPost and base vs. pedestal and fork

Tonight, I got to do an experiment that I've wanted to do for some time now.

For years, I've heard in conversations with people who shall remain nameless (unless they care to contest this work) that the 3/4"-post-on-rectangular-1/4"-thick-base optic support method (used at the 40m and adopted into LIGO) is better than the standard 1"-pedestal-and-fork method (used by many experiments in our own labs and elsewhere). After many attempts, I have never succeeded in getting any hard data to support that claim. So, I decided to make a measurement myself.

I set up a simple michelson using one of the SiFi beams, once using each support scheme for the beamsplitter and end mirrors:


There is a HWP to find the polarization for which the "50:50" BS is closest to balanced, a lens to focus into the IFO, a second lens to focus the AS beam onto the PD, and the PD itself, which is a PDA255. There is an ND=0.7 filter on the PD.

Below is a scope screenshot of some fringing action when pushing on an end mirror. The contrast defect in each case was pretty low at ~2.5 x 10-3.

Once the IFO was aligned in each case, I pushed on one mirror a bit to creep it into a half-fringe state. This took some time, since I had to push and wait a few seconds for it to settle. After doing that, I took a spectrum (actually 3 at different spans from 1 kHz to 100 kHz). The results are below, with a zoomed plot to the right.



As you can see, the difference is pretty minimal. The post-and-base setup has slightly higher RMS below ~1 kHz, owing to two high-Q resonances (at 340 Hz and, to a lesser extent, 920 Hz). My detractors will accuse me of bias (e.g., in tightening, etc.), but I invite anyone to come test this with their post-and-base clamping chops.

With this, I'd like to put to rest the notion that the post-and-base method is somehow fundamentally superior. I DO acknowledge that there are definitely wrong ways to use the pedestal-and-fork, and this can lead to the non-idealities noted in the folklore. The post-and-base method is foolproof in a way, since the proper procedure is somewhat manifest (use two screws, use washers, etc.), while the pedestal-and-fork requires some diligence to get just right. However, with just a little bit of care up front, the pedestal-and-fork offers huge advantages:

  • Arbitrary placement (most importantly, the ability to always work along table hole axes, which makes alignment incredibly easier).
  • Free-hand alignment, since the fork can be placed without jostling of the optic (see below).
  • Space, since it takes up way less of it.


How to fork

To mitigate the potential recklessness of this post, I offer the Zach-Approved™ Forking Method.

First of all, this is the only fork you should be using, the Newport PS-F (maybe there actually are other acceptable ones, but none that I've found that don't apply horizontal forces on the pedestal upon clamping):

Now, the forking method (accompanied by the GIF below):

  1. Locate where you want to place the optic.
  2. Place the optic and align by hand. The pedestal is heavy enough to hold itself in place with friction.
  3. Locate the appropriate screwhole and set the fork gently onto the pedestal, slightly un-engaged. Of crucial importance is that you pick a hole that will be somewhere close to the middle of the slot. Working along the table hole axes ensures there should be several options in every case.
  4. Using your finger, engage the fork on the pedestal. If you use just enough force to move the fork, it will come to a stop when fully engaged and you won't have moved the pedestal (and optic) at all.
  5. Insert the screw and hand-tighten a few threadlengths.
  6. Use a ball driver to finish screwing almost until tight.
  7. Just before tightening, give the fork one final nudge against the pedestal. Sometimes I use the ball driver tip, but this can also be done with your hand so that you can keep the driver in the cap.
  8. Tighten as desired.

At no point after initial alignment should you have to touch the optic, and, if you follow the procedure above, the optic orientation should not have shifted by more than a mrad or so. You can see how little the optic moves over the operation in the GIF.

  1237   Sat Apr 11 19:03:58 2015 ZachElectronicsSensorsRL readout PD

Seeing Hartmut's talk at the last LVC meeting about innovative DC photodetector designs (something necessary for future squeezed IFOs) reminded me of some investigation I did into the same while at LLO. One thing I did a fair bit of work on while there was the DC current subtraction idea (c.f. LLO:6449 and 6532), but another thing I spent time modeling was the concept of using an RL network, as Hartmut is exploring now.

The circuit I was considering differs somewhat from Hartmut's idea. In his circuit (at left below), the inductor ("L1") and input resistor ("R1") perform a current branching: at low frequencies, the photocurrent sees low impedance to ground through the inductor, and therefore does not pass through the transimpedance amplifier and get converted into an output voltage; at high frequencies, the inductor looks like an open circuit, and all the current passes through the TIA. Ideally, this leads to an effective frontend whitening that allows for a high Z at audio frequencies. In practice, one would use either the DC resistance (DCR) of the inductor, or perhaps an extra resistor in series, to set the DC Z, which would be Z(0) = R_2 \frac{R_{DC}}{R_{DC} + R_1}, where RDC is the DC resistance of the inductor path. One problem with this design is that, since RDC cannot be arbitrarily low due to the DCR of the inductor, one must choose an R1 that is high enough to set the DC Z to a low enough level. Roughly speaking, this means that the value of R1 must be approximately the ratio of the desired AC and DC transimpedances (typically a factor of 100 or so), times RDC. Since RDC will be on the order of 100 Ohms, R1 must be on the order of 10 kOhm. This in turn means that the current noise of the amplifier is fully converted by this high impedance at all frequencies, which ruins the SNR of the detector at low frequencies (you could use a low-current-noise part, but then the voltage noise kills you directly).

The circuit I had in mind is at right below. As you can see, the amplifier in this case is only used as a unity-gain buffer for the passive readout circuit (though one could consider adding a switchable flat gain for low-current operation, as in the ZSWITCH feature of the currently used DCPDs). This design works simply by having a passively different transimpedance at different frequencies: at low frequencies, the inductor shorts the large resistor and the transimpedance is just the inductor DCR plus the additional series resistance to ground (50 Ohms in the schematic); at higher frequencies, the impedance increases until it is limited by the parallel resistance (10 kOhms here, plus the series 50 Ohms). With this topology, the current noise always sees the same impedance to ground as the photocurrent does (i.e., the transimpedance), and there is no extra reduction in SNR. The "DC" section is not necessary in principle, and in fact it always has worse SNR for a finite inductor DCR, but it could be used as a calibration path for the DC response due to potential nonlinearity of the inductor.


As a side project, I've started doing some testing of this design. To start, I bought a ginormous 4-H inductor from DigiKey:

Transfer function

The first thing I did was to verify the transfer function. To do this, I biased one of our 3-mm diodes with the M2 circuit bias supply, then sent the anode into a breadboard version of the RL circuit. The parameters were slightly different: L = 4 H (DCR ~ 60 Ohm), RAC = 10k, RDC = 39 Ohm. I then put one of our SiFi lasers on it and modulated the power using its fiber amplitude modulator. Here is the result:

As you can see, it performs just about as expected. A couple notes:

  • The LISO trace above has been adjusted using the inductance and diode capacitance as fit parameters, since they are not known precisely a priori.
  • The slow upturn at low frequencies is fairly well explained by the amplitude modulator response at low frequencies (see CRYO:1187).


Of course, the biggest concern with using such a big inductor is the additional noise it might inject, particularly due to pickup. Below is a summary plot of some measurements I made on this circuit, together with some theory curves and the currently used DCPD for comparison.


  • Obviously, there is a strong presence of pickup here. These traces are also the result of wrapping the inductor in metal and orienting it to minimize noise. Yes, the pickup is prohibitively bad as it appears here, but I have an idea to get rid of it (see below).
  • Below a few Hz and above ~1 kHz, the circuits behave as they should, except that the LT1128 exhibits excess current noise in the high-frequency region. This is no surprise, as we have never acheived the "typical" noise performance advertised by Linear (see ATF:1890---Rana points out that this could potentially be due to our not following proper electrostatic discharge practices, but it seems that they are always out of "typ" spec---not "max"---in the exact same way). In this case, though, the noise is even worse than the "max" level, which is an extra party foul. My proof of this is that the OP27-stuffed circuit does what it says it should at these frequencies. Note also that I had only some modest thermal shielding, and that's why the noise shoots up below a few 100 mHz.
  • Given the typical LT1128 performance in the real world, it may be that the OP27 is the ideal buffer amp for this circuit, since it adds a factor of a few at high frequencies (where we are most likely to be sensing noise limited). It's also only a factor of a few worse than the fantasy LT1128 curve at low frequencies.
  • For all its nascent imperfections, this is still already a detector with 10x better noise both at 1 Hz and above 1 kHz than what we have in aLIGO, so that's already promising. yes

Reducing the pickup:

So, we are left with the problem of being highly sensitive to a signal injected into the coil, but not to one induced by external fields. Luckily, the guitar industy has had a solution for almost exactly the same issue for about 80 years now: the humbucker. Of course, it's slightly different, since in the guitar case you want to be sensitive to an induced signal (i.e., the signal from the string, which is deliberately made differential-mode by reversing the polarity of the magnets inside the coils), but I believe the same principle should apply. In our case, we'll put the inductors in series electrically, but adjacent and flipped spatially. In that case, the pickup-induced voltages should cancel while the current-induced voltages should add, as desired. It's hard to find CMRR values for high-end humbuckers, especially since they are usually intentially imbalanced for tone considerations, but I would venture a guess that the ~40-50 dB required in this case is not completely out of the question. I've ordered another identical inductor to see what we can do.

  1236   Wed Apr 8 21:12:19 2015 DmassCryostatSiFiRe-evacuated large cryostat

You can make them yourselves if you want with activated charcoal and stycast - you just need to bake and pump to clean it. I got some activated charcoal from Keith a while back (it's either in the glass cabinets in the Cryolab, or the fume hood in the ATF), but I think you'll need to buy some new stycast (which you should do anyways if you're going to be gluing stuff in there - the stuff we had for the Cryolab is past its expiration). The cans of what you want are in the ATF fume hood.

If you have room somewhere by a screw hole, you just make machine a mounting block, cover it in HV compatible epoxy (stycast), and roll it around in activated charcoal. To clean it, just pump and bake (the setup to do this for this should be in the CTN lab, if you cannot figure out what it is / how to use it and are nice to him, Dmass will probably help you). If you do it this way, you can unscrew it and pump/bake it when it loses its vacuumy goodness/cleanliness.

If you don't want to keep vacuum but don't want to ruin your getter, just backfill with N2 gas the valve off. Keeping the cryostat open will unavoidably dirty it up though.

  1235   Tue Apr 7 16:24:18 2015 ZachCryostatSiFiGetter quote

Apparently, the getter is made by IRLabs themselves. They've quoted me $690 for a replacement (see attachment *removed since I was informed that's not allowed*).


Good point. If you or Nic can find it, please post some getter info and we can get a spare just in case this one is already stuffed.


  1234   Sat Apr 4 18:08:18 2015 ranaCryostatSiFiRe-evacuated large cryostat

Good point. If you or Nic can find it, please post some getter info and we can get a spare just in case this one is already stuffed.


I roughed our cryostat back down tonight. Remember, this model has a getter that degrades over time when exposed to air, so we should minimize the time that it's not under vacuum.


  1233   Thu Apr 2 20:21:18 2015 ZachCryostatSiFiRe-evacuated large cryostat

I roughed our cryostat back down tonight. Remember, this model has a getter that degrades over time when exposed to air, so we should minimize the time that it's not under vacuum.

  1232   Thu Apr 2 01:03:33 2015 ZachLab InfrastructureSiFiRack set up, table cleared and real layout started

Today, I installed the shelves into our rack and moved the electronics that were on the table into it. I then cleared the table and started setting things up the way they will be.

The cryostat is now on the south side of the table, and the lasers are set up on the north side. I connected and placed the AM and PM modulators in both laser chains, then placed the output couplers. I used posts to protect the fibers and strain releif the SMA cables going to the modulators, and installed the BNC patch bay on the end of the table near the rack. I'm waiting on our 3/4" posts to continue with the optical layout.

With everything connected, I trimmed the AM DC voltages to maximize the transmission and verified that the output power for each was ~25% what it is directly out of the laser, as it should be (since each modulator has loss ~3 dB).


  1231   Mon Mar 30 17:56:13 2015 DmassNoise HuntingNoise BudgetWest Photothermal Transfer Function Actually Measured

Loss vs T. Confused on error bars here, so ommitted for now. This will have error bars before I thesify it.

I used the transfer function as well as our knowledge of the noises to estimate the coherence of each meeasurement, and then that to estimate the fractional uncertainty of the (SR785's estimate of the) transfer function, but the fractional error was significantly smaller than the residuals (e.g. the systematic contributions to the reduced Chi^2 dominate over the random noise).

Error bars can still be derived from the curvature of the residuals over the fit parameters.

I tried to use MATLABs curve fitting package to give me error bars from variance over the measurement, but couldn't figure out how to give it the farsi code results in a way that it would fit over.


I spent a while trying to measure the PT TF @ 125K and 114K, but:

  • The 125K measurement point is all noise for unknown reasons (see elog:1226 for raw data)
  • The 114K measurement baffles me - I think the optical offset coupling (flat) and the PLL offset sensitivity are crossing, but they cross with a slope difference of f (f^0 and f^1), and notch. I thought about it for a while but do not know what could cause this phase disparity.

If I spend a while (~half day? day?_ longer on fitting with the 114K measurement, I might be able to add a point to the plot. Time is precious right now so I am tabling it.

Given the uncertainty in why I couldn't measure the TF cold, I am tabling the idea of having another cooldown to get more data until after I defend. The best scientific statement I have right now is:

"The results indicate that further exploration of the transfer function is warranted*"

*if anyone cares about the results.


I took cavity pole measurements (via transfer function)

  1230   Fri Mar 27 00:26:02 2015 ZachSummarySiFi - ringdownTaiwan cantilever phi vs. T (old but unreported)

Before the LVC meeting, I had just done a long steady-state Q measurement on the Taiwan cantilever. I got too distracted by the melted epoxy disaster (CRYO:1225) to actually post the data.

Below is a plot of the ~16 hour stretch of data (second trend), showing the temperature and instantaneous loss angle. The temperature was stepped in 10-K increments from 90 to 130 K, holding at each temperature for 3 hrs to allow the system to equilibrate and integrate (except for some of the early steps which required some manual intervention).

The main result is that the loss seems relatively constant at ~10-6 from low temperature to ~120 K, where it starts to increase. Towards the end of the 130-K stretch, the LN2 ran out, and the system started heating to room temperature uncontrolled (i.e., heater output was railed at zero).

This level is too high to be from the Si, so I assumed it was some residual clamping loss. I was dubious that the figure from the one reference that Matt A. found and gave me for the cryogenic Q of stainless steel would be applicable to our particular clamp, so I thought I might try to measure it directly, in parallel with the cleanup of mess in the cryostat. To do this, I got some spare steel wire from Gabriele and made a makeshift suspension, hanging the top piece of the clamp, hoping to measure the loss of its lowest vibrational mode. I knew it was a long shot, since this mode should be around 17 kHz, but I set it up in the simple vacuum chamber anyway, and tried to excite it and read it out optically. The first bending mode should have nodes *near* the suspension points, so I thought I might get some kind of meaningful results if I could actually see a ringdown.

I was unsuccessful. I tried various excitation schemes, from broadband (banging stuff) to narrowband (bandpassed white noise, amplified with the boom box and blasted out of a speaker touching the chamber), and none revealed any mode excitation. I was able to see broadband noise increase with the excitation profile, but no lines, so most likely I was seeing some alternate path.

I still think it would be nice to get an empirical measurement of the cryogenic Q of the steel we use for our clamps. Maybe we can set up a laser vibrometer measurement like Norna and her student did a few years back on the steel gyro PMC?

  1229   Thu Mar 26 20:06:02 2015 ZachCryostatSiFi - ringdownSmall cryostat reassembled, Taiwan cantilever in clamp, pumping down

[Den, Chris, Nic, Zach]

Since my snafu before the LVC meeting (CRYO:1225), the small cryostat has been in pieces being thoroughly cleaned and aired out. Nic wanted to have the ringdown setup rebuilt so that we can demo the steady-state Q measurement technique for our visitors, so we did some work today to make that happen.

This morning, I re-lined the main chamber walls and floor with aluminum tape. This model came with some thin foil lining the walls, attached by periodic thin strips of double-sided paper tape. We have been intermittently scraping some foil off each time we cycle, and since a nasty residue was present on the floor of the chamber after the epoxy incident, I figured it was time to replace the lining. I just used aluminum tape since a.) it is stronger and will be less prone to scraping off, and b.) if and when we need to replace it again, it should come off much more easily.

This afternoon, we rebuilt the cryo package on the cold plate (clamp with Taiwan cantilever installed, ESD, and 45º mirror). Since we don't want to use epoxy to mount the power resistor anymore and we don't have any tapped holes in the clamp, we have not equipped any heat source or temperature sensor. This is fine, since we really just want to use it as a demo this time around, and room temperature should be sufficient. If we want, we can still cool it down to LN2 temperature, but we won't have any actuation or readout.

Upon pumpdown, we noticed that the pressure had stalled at around 20 mTorr after a good 20 mins of pumping, indicating that we had a leak. We checked the top seal and electrical feedthrough (which had also been freshly reattached during the rebuild), and found no issues. With nothing else to try, we decided it was most likely the seal between the chamber floor and the main section (I had to foil this with rectangular sections of tape, which I then XActo cut into a circle at the o-ring groove, so it was possible that a foil flake was blocking the seal). With everything still in place, we flipped the cryostat over and removed the bottom. We found a couple places where a tiny piece may have extended into the seal, so I re-cut the circle more conservatively. When re re-sealed, we found the pumpdown profile to be much closer to what we usually expect. The pressure was a few mTorr after ~10 minutes and showed signs of healthy decline.

We rebuilt the optical readout, then tested the MODERINGER amplitude sensing and found everything seemed to be working. We did not want to test the ESD at this high pressure. When I left, the Q was relatively low at maybe a few thousand, but gas damping was likely still a limiting factor. Also likely is that there is still some residue on the cantilever that I didn't get off, or perhaps even that some irreparable damage might have been done. We should be able to tell when the pressure is low enough.

ELOG V3.1.3-