I removed the 10 dB attenuator between the LB box output and the laser driver's current modulation input. The ranges of the LB box output and ITC 502 input are matched, so there should be no need for an attenuator. I also slightly reduced the servo gain to 5.3. There is still a 10dB attenuator between the PDH mixer/lowpass and the LB box input, to avoid saturating the servo's input stage (+- 330 mV).
After moving most electronics to the same power strip as in the above elog and tying the table to ground, I no longer see a 144 kHz oscillation in the PDH error signal (oscilloscope). I do see a ~40 Hz oscillation in the PDH error signal... could that be the fundamental mode of the cantilever? Zach's thesis quotes a fundamental resonance at 45 Hz. The 40 Hz oscillation is present even while unlocked (but with the servo engaged).
Though the error signal looked clean, the lock was not steady for more than 10s of seconds. I suspected that the current control loop was far from the center of its range. I am monitoring the current control signal from the LB box Aux output, which picks off the signal before the final summing amplifier stage. Therefore, in order to adjust the temperature setpoint to 'zero' the control signal, I needed to first disconnect the LB box inputs then adjust the output offset until the voltage at aux output matched that at the final output. Then, with the laser locked, I adjusted the temperature setpoint until the control signal as measured at aux output was zero (plus some offset that was present at both the main output and aux output even with the inputs disconnected).
After adjusting the temperature setpoint manually, the beamspot is noticeably brighter on the trans mon camera, and the lock is stable for minutes.
I tried setting up the digital control to handle the temperature loop, but the realtime system is in a bad state and I didn't attempt to diagnose it.
I traced the fiber beam path and confirmed that the system is set up as in elog 2833. I add a FC/PC-to-FC/APC patch cable between the 50-50 beamsplitter and the 1611FC-AC photodiode. We've been meaning to add that patch for a while to avoid damaging the connection.
I opened the connector housings on the cable for the N laser TEC and readjusted them such that the internal clamp bites down on the jacket of the cable, rather than closing over the conductor wires themselves (see attachments 1 and 2). Wiggling the cable no longer causes an alarm on the TEC.
I noticed that the FC 1611 photodiode was being powered from a GlobTek +-12 V DC supply with 0.25A current limit; however, those photodiodes require +-15 V DC with 0.3A current limit. I located the proprietary cable and am now powering that photodiode from the New Focus 0901 power supply located on the PSOMA rack.
I also noticed that the SMA cable from the fiber box panel to the spectrum analyzer that should have been carrying the beat note was connected to the wrong port! That's probably why it wasn't visible before.
With those changes, I could see the beat note on the spectrum analyzer .
I wanted to add a cable from the beat note PD DC output to the fiber box panel, so I could check that the DC power on that photodiode makes sense. But, I couldn't locate one in cryo or EE. Maybe the 40m or another WB lab has a spare?
labwork to do
I powered on the 'North laser' using an ITC 510, and make set up the fiber optics for the beat note as depicted in Attachment 1:
I'm collecting useful calibration/gain parameters in Elog 2830 for quick reference.
Shruti has observed a 144 kHz oscillation on the PDH error signal from our cavity-with-cantilever. The oscillation was railing the PDH signal, making it impossible to maintain lock for more than a few seconds. I came in to troubleshoot.
That's all for me this morning. I think the oscillation is sufficiently low we could try using the DAQ to feed back to temperature as we were before. It would be useful for diagnostic purposes to maintain a more extended lock, and I'm finding I need to tune temperature anytime I reacquire lock. Maybe we're just always sitting close to the edge of the current control loop.
Videos of flashes and a very noisy lock.
In Attachment 1 and the video showing a lock, the PDH signal essentially goes to its rails. Most of the noise is in oscillations that are roughly at 143 kHz.
This elog (2815) shows the cantilever with its mount and a top view of the cavity.
In order to improve the acoustic isolation of the box I used rubber tubing cut as depicted in Attachment 1 and attached to the edge of the plastic box (as seen in Attachment 2).
Attachment 2 also shows one of the four dog clamps I used to secure the box. Finally I placed an oscilloscope on top of the box to provide some weight.
The box has 4 holes for the input, reflected, transmitted-to-PD, and transmitted-to-camera beams.
In elog 2731, the PDH signal is almost 1 V pk-pk, but recently I had only been seeing a very noisy signal ~10 mV pk-pk (purple trace in Attachment 4. Transmitted light is yellow, reflected light is blue). I adjusted the lengths and removed the attenuators in the path of the EOM but it did not seem to change.
While going through the path of the cables once more, I realized that signal from the OCXO was being sent to the wrong port on the rack mount (Attachment 5). It was earlier on D3, and I later changed it to A4 which was labelled 'S EOM', but the actual location was D4 which I identified from this diagram. Now I see a reasonably large PDH signal as expected with ~800mV pk-pk (video here).
Since clamping the cantilever, I began to align the triangular cavity with the cantilever.
Attachment 1 shows the setup of the cavity with the input coupler at M1, curved mirror at M2, and a cantilever with a flat optic at M3. The cantilever does not possess any knobs for fine adjustment and its initial mounting is fixed. M1 and M2 can be adjusted, but to get to TEM00 resonance, and any fine alignment, the two steering mirrors before the cavity were used.
1. While aligning, I began to see resonant peaks corresponding to the cantilever's mechanical resonance showing that the cantilever has a very high Q. I mistakenly thought the cavity was close to being aligned to resonance in this configuration while also being very stable, but later understood that the strong oscillations actually seemed more like accidental misalignments creating an optical lever. The green traces in the videos are the measured transmitted light through the cantilever. When the cavity is really aligned close to the TEM00 mode the transmission looks like the green trace here, and the reflection (from M1) is the blue trace.
2. Another time I thought it was aligned I was looking at the camera at the transmission of the cantilever [M3 in Attachment 1] and could see a single flashing spot but when I moved the camera around I actually ended up seeing two beams instead, simultaneously flashing. Minor changes in the alignment resulted in the beams individually and simultaneously turning into higher order modes. Initially I thought that the situation was as depicted in Attachment 2 and found that it was a consistent geometric solution, but later Koji pointed out that was the geometry that creates the TEM01 mode. In Attachment 2, the initial beam (red, M1->M2) overlaps not with the beam in that path from the second round trip (yellow, M1->M2) but with the one from the third round trip. Was it possible that this misalignment was too severe that instead of being TEM01 two separate, but identical, resonant cavities were created? A measurement of the FSR, which I did not do, would have proved that that was the case.
Since the transfer of fiber components into the box, I have re-aligned the cavity to the TEM00 mode. Note: replacing the fiber at the fiber launch and also changing the polarization at the fiber launch moves the beam around.
[more information and data coming soon]
1. The PDH signal seems too low, possibly because the LO and RF relative phase must be re-adjusted.
We've been learning the various processes required to fabricate silicon cantilevers following Zach's recipe.
The last step involves etching cantilevers out of the silicon wafer using KOH. Specifically, Zach's etch recipe for a 500 um wafer coated with 400 nm SiNx on both sides is:
I tried to follow this recipe last Wednesday, with the following modifications:
Because the wafer is 3/5 the thickness but the etch rate is also 3/5 as fast, I anticipated that a 6 hour etch would be appropriate to produce something for scribing and breaking... however, after 4.5 hours, the entire wafer was almost entirely dissolved. All that remained were thin, fragile sheets of Si or SiNx. What's going on? Some possibilities
I have since perused the KNI wiki for more resources on etch rates through Si. The most extensive data seem to be from the BYU page on KOH etching. It suggests that 30% KOH at 80 C etches through 100 Si at 80 um/hour (compared to 60 um/hour that Zach was using), while 50% KOH etches through 100 Si at 45 um/hour. This doesn't really explain the results. Even at 45 um/hour, we should have been left with 100 um of Si after 4.5 hours, or 1/3 of the initial material. If we take Zach's 60 um/hour at face value and apply the data from BYU as a 'relative rate of etching,' we would be scaling by a factor pretty close to that suggested in Silicon Micromachining.
I wasn't able to find good data on the rate of KOH etching through SiNx depending on temperature or concentration.
We are using doped silicon, but I found reference online to boron doped silicon etching more slowly.
The bath temperature could also have been systematically higher than 80 C, or just not well controlled around 80 C.
To test this, I'm planning to do a shorter etch of a sacrificial piece of Si under the same conditions as before (30% KOH, 80 C). I'll remove any oxide layer with HF beforehand, and check the bath temperature with a thermometer.
I used the standard SiNx deposition recipe on both sides of the wafer, but did not check the resulting mask with an ellipsometer. We should in the future do this after most previous steps: PECVD, optical lithography, DRIE etching, etc.
I'm being trained on the ellipsometer this afternoon, and plan to measure the thickness of SiNx on some 4" wafers Zach had left over.
Our cleaning process involves a few minutes in an HF bath, and we use DRIE etching during the optical lithography step. Either of these processes could thin the wafer. In particular, during lithography I noticed that my photoresist was a bit thinner than I'd intended. Perhaps the I etched through the exposed photoresist more quickly than anticiapted, allowing the DRIE etch to reach the underlying silicon for longer.
In the future I'll measure the thickness of the edges of the wafer (where there is no cantilever) with a micrometer before etching.
This afternoon, I tried to measure the etch rate of the KOH bath. I did the following:
I had intended to remove the wafer and measure its thickness again, so unfortunately can only place a lower bound on the etch rate. Nonetheless, the implied rate of etching is >16.5 um/minute (etched through at least 290 um in 35 minutes, from both sides of the wafer). This is more than an order of magnitude faster than expected, even allowing for the increased bath temperature. Clearly I am missing something -- is the KOH actually being diluted due to some additional DI water in the pumping system? Is the boron doping really increasing the etch rate by that much? Did the wafer just fall off of its holder and get lost in the murky KOH (I did fish around for several minutes, no sign of a wafer in the bath)?
I mounted the fiber components (north and south Rio lasers, Faraday isolators, 90-10 beamsplitters, 50-50 beamsplitter, fiber EOM, 1611FC-AC) inside our breadboard-in-a-box from Thorlabs (attachment 1).
Along the way
Though I set up the beamsplitters in the box approximately where we will eventually want them, the beam currently does not pass through any beamsplitters. I left the optical layout for the south laser identical to how it was on the table, with the addition of a patch cable between the EOM and the launch: laser -> faraday -> EOM -> patch cable -> launch. The north laser is not connected to its drivers.
Yesterday while the power was out I turned off laser drivers and other powered electronics that I could think of. When the power was back I rebooted the computers (checked that I was able to ssh into cymac) and the electronics that I had turned off. I noted that the cavity was back on resonance as indicated by the forest of peaks that we were seeing.
Today when I returned to the lab the forest was missing and Aaron noticed that the cantilever seemed to be tilted in the clamp. Attachment 1 shows the exact angle at which it was found. A small part at the end is chipped off but it otherwise appears usable to me.
Attachment 1: The cantilever at the angle at which it had tilted to with the chipped bits, after just removing the top piece of the clamp.
Before re-clamping, we (Aaron and I) decided to use a variable torque socket wrench to test for optimal clamping torques on another sacrificial cantilever. I decided to abandon this experiment after Rana pointed out at the group meeting that this would not give any really reliable number for the required clamping torque since breakage torques may have a wide range of random values for different cantilevers.
Attachment 2: Mobile phone image (without external lens) of the cantilever's clamping surface. It shows some pits and scratches, one larger than 100 microns.
For reference, the dimensions of the clamping area are 1cm x 1cm.
Attachment 3: The clamping surface of the smaller (top, while clamping) piece of the clamp, that contains the indentation where the cantilever sits.
Aaron removed the plastic wrap and began powering and reconnecting all electronics.
I drilled three holes into the plastic box that covers the cavity, reconnected the south laser diode and TEC to the ITC 502 combi controller and set the laser temperature to 9.888 kOhms and current to 126 mA (until the beam was visible).
Continued with alignment of the cavity until I observed a forest of peaks again while sinuoisoidally oscillating the temperature, although they can be observed even without temperature cycling now that we ahve a cantilever.
[aaron, shruti, raj]
We added the acrylic framing that supports the HEPA FFU, as well as two of the 3 panels covering the top of the enclosure. Raj and Shruti also unwrapped and added handles to the orange acrylic doors.
However, the last acrylic panel for the roof of the enclosure is too small. Though the part is cut as specified at 23" on the short side, the counterbore holes on the long side of the panel do not span the distance between the two rails where the panel mounts. The counterbore holes are 0.75" from the edge of the panel, which means the two lines of holes are 21.5" apart. However, the distance between the center of the rails is specified at 22.5". The distance between the rails is appropriate for the size of the HEPA FFU, and the 0.75" gap along the edge of the panel is standard across the other roof panels. Therefore, I suspect the overal width of the panel was just specified 1" too short.
I'll alert F&L and have them send another panel ASAP
[aaron, shruti, chris, raj, radhika]
On Tuesday, Shruti and I did fit checks of all connectors for the enclosure. We received the remaining parts according to the parts list from F&L, so should have everything we need. We requested clarification on where to use a few connectors, though haven't yet received a reply (our contact at F&L was unavailable).
With some adjustments, we completed steps 1-5. The frame of the enclosure is around the table. Tomorrow we'll try to complete the rest of the build, which includes adding the roof and panel doors to the enclosure.
Backups were restarted for the cryo lab computers gaston, spirou, and cominaux. A 4TB USB drive was connected to cominaux, mounted under /backup, and rsnapshot was configured to run on a nightly basis. It does not back up the full disk, but only those directories where user-generated files are kept (/home, /etc, /usr/local, /opt, /ligo). rsnapshot's configuration files are: /etc/rsnapshot.conf and /etc/cron.d/rsnapshot.
For the cymac, configuration and minute trend files are backed up by rsnapshot to a 1TB disk, mounted as /backup on cymac1.
[shruti, aaron, raj, chris, ian]
I've asked F&L for a parts list so we can do a proper inventory before starting assembly. Also requested a quote on purple panels, update on shipment for the cross bars, and imperial threaded feet.
Shruti and I transferred the materials for our new enclosure from receiving to the subbasement hallway. I noticed that the acrylic panels we received are amber, not purple as we specified... unfortunately it looks like the error was already present in the quote, which listed acrylic #2422 (in my emails with their rep, we'd both confirmed acrylic #2424). Because it was in the quote, I'm not sure there's anything we can do, but I'll ask.
Ian, Raj, Shruti, Chris, Aaron, and possibly other grads will meet at 10 tomorrow for assembly.
We returned CFC-2X-C into the same cabinet it was borrowed from. Thanks CRYO!
I've borrowed 2 collimating lenses (for fiber input) for use with the free-space AOM in the DOPO lab.
Today I began to see a forest of peaks in the transmission (transmission through the cantilever optic) while aligning meaning that the cavity is around resonance. I then adjusted the Watec camera to the transmission of the curved mirror and began to see some flashes. I took a video showing this which can be found here.
I tried activating the feedback loop with the settings we had on earlier but while it did seem to increase the peak powers of the peaks it did not seem to lock. The steering mirror to the refl photodiode (Newfocus 1811) and the corresponding lens needs to be re-adjusted.
Similar to elog 2767, before changing out the mirror from a rigid mounted one to the cantilever, I measured the loop transfer functions and noise spectra since we had not done so since we moved the table.
The only change between the current and previous version was that I tried to make the low frequency phase offset zero.
Attachment 1 shows different estimated open loop transfer functions corresponding to ratios of specified closed loop transfer functions (as mentioned in the labels). The UGF was lower than what was previously acheived.
Attachment 2 shows the fit of the slope of the open loop transfer function as done previously in elog 2776.
Unfortunately, I did not notice the issue with binning when I measured the noise spectra which resulted in a discontinuous spectra corresponding to the different regions I measured separately on the Moku. The data for this and everything else mentioned in this post is in Attachment 3.
In short: Between Friday and today, I de-bonded the mirror from the broken cantilever, picked two cantilevers and bonded both the mirrors to them with the AR surface being the one on the bond side, clamped and mounted on a post one of these cantilevers, and placed it in the cavity replacing one of the flat mirrors (M3).
For this, I used methanol and soaked it for around 30 min similar to the procedure here. Since this bond was weaker, it took lesser time.
I picked one cantilever from the previous collection and another from the dish shown in Attachment 1. This second cantilever did not seem to have the back surface passivated with SiNx and looked like it was just oxidized silicon.
I used the GE cryo-varnish at four points around the mirror placed on the etched edge of the cantilever for both and let it dry for a few hours (Attachment 2). The bonding was performed such that the AR surface was on the side of the cantilever.
Using the alignment jig on the table, I picked the cantilever without the SiNx passivation on both sides (identifiable as with one non-glossy side) to mount to the clamp. I tightened it not too strongly and then used a 3/4" post of suitable height to get the center of the mirror to a height 4" from the table.
I placed it roughly where I thought it should be on the table replacing the flat mirror M3 from the previous setup. I got the beams from two round-trips to overlap visually and added a PDA10CS to look at the transmission from the cantilever.
The aim now is to do as we did earlier - sinusoidally change the temperature by driving the TEMP TUNE input with a function generator and slowly tweak the alignment of the steering mirrors to find the cavity resonance.
I also measured the transmission of the cantilever as roughly 0.1% using the power meter.
I broke the cantilever while fastening the fork clamp. Afterwards, I used methanol to remove the varnish and separate the mirror from the cantilever shard. I then used a cotton-tipped swab soaked in methanol to clean the varnish from the sides of the mirror. I drag wiped the HR and AR surfaces of the mirror with methanol followed by isopropyl alcohol. Finally, I bonding the mirror to a different cantilever (this one with somewhat more pitting than the previous) -- again bonding with cryo varnish at four points on the sides of the mirror, but on recommendation from Chris this time with the mirror AR surface touching the cantilever.
This morning, I clamped the cantilever-with-mirror that we bonded yesterday.
I first used a broken cantilever to practice the clamping. I clamped the test piece first in the nominal position, then with slight alignment errors in the available degrees of freedom to see what those errors would look like. If the cantilever is misaligned such that it does not rest in the clamping groove, the clamping block will show a larger than usual gap on at least one side (attachments 1, 2). If the cantilever is not entirely 'in' the clamp, the thicker part of the cantilever will be visible above the clamp (attachment 3). Attachment 4 and 5 show the test cantilever in good alignment.
Next, I clamped the cantilever-with-mirror by
The Q may improve with more clamping force, but I'd like to do a more controlled test of how much torque can safely be applied to the clamping screws. We don't want to break any usable cantilevers.
[Shruti, Koji, Aaron]
Today Koji guided us to (1) remove the mirror I had contacted to the cantilever that later broke, and (2) clean and contact the second flat (1550 nm wavelength coated) mirror to another cantilever using cryo varnish.
After my failed attempts at using methanol and acetone along the edges followed by soaking the contacted-to-silicon optic in isopropanol, today we were finally able to de-bond the mirror by soaking in methanol. We used two washers to raise the optic from the aluminum surface and folded foil pieces to keep everything in place. The setup shown in Attachment 1 was covered with another aluminum foil dish and let sit. It took at least an hour of soaking to completely de-bond.
Koji then drag wiped the optic with pure isopropanol to clean any remaining residue and cryo-varnish hairs.
First we tested drag wiping the broken silicon cantilever with methanol and isopropanol. Then we selected a reasonably looking cantilever that was not chipped at the edges, though it had an uneven surface, and drag-wiped both surfaces with isopropanol.
We drag-wiped the HR surface of the optic a few times and then placed it on the portion of the cantilever that had the square etched-out region. While Koji held the mirror in place I applied cryo-varnish to four points around the mirror. This is now set to dry.
Update: Photos are available on the ligo.wbridge google drive. I uploaded everything, but could pare down to save drive space.
I borrowed this free-space AOM (1550 nm) for use in the DOPO lab.
The cantilever clamp with a cantilever on it to be used on an optical post. This will replace the non-transmitting flat mirror in the PSOMA cavity. There are two screw holes to mount the optical post, but it would only be possible to use one, therefore this will be initially mounted asymmetrically.
Cantilever with the two flat mirrors from Zach's setup. Without having it properly mounted, I could only very crudely measure its reflectivity as R>81% using a power meter.
The one on the left is shown with (what I believe, by comparing to the curved optic in our present setup, is) its HR surface facing up and the one on the right with its AR coated surface shown.
I attempted to bond the mirror to the cantilever with cryo-varnish after testing that it bonded two pieces of silicon. The HR surface which would be facing the interior of the cavity was the side I decided to put the varnish on because the pictures from Zach's thesis depicted the cantilevers that way. I coated the edge of the square-shaped hole on the cantilever head with the varnish and placed the mirror on it. I placed a lens wipe on top of the mirror and then another flat optic over it. While applying pressure on it with a tweezer the cantilever broke (possibly because the aluminium foil below it was very crinkly which I didn't think too much about beforehand).
While the mirror seems to have bonded after a couple of minutes of holding it down, I probably have to remove the varnish (with IPA/acetone I think ?) and re-bond to a different cantilever.
What about showing the E/W beat noise for several different operating points on the W laser? Temperature and current.
Between 0.5 and 50 Hz, there are a couple of regions where the Rio W laser noise dominates the three corner hat measurement. And, below 0.3 Hz, the Teraxion laser noise dominates the measurement. Today I'm going to try to quiet these two lasers a bit to make a slightly improved three corner hat estimate.
I'll be looking at the output of the delay line frequency discriminator (DFD) on the moku spectrum analyzer, so I've swapped in the 1.9 MHz lowpass filter for the one Rana was using to check the noise out at MHz.
I saved the following traces on an SR785, but with a nonfunctioning GPIB so ended up storing them on a floppy drive.
I saw minor improvements to the Rio W x Teraxion beat note spectrum, and took 20 minutes of data for each beat note. Dropbox upload failed several times, so I sent it to my laptop via ipad file storage and airdrop. After swapping fiber connectors and covering with foam, the system took ~15-30 minutes to equilibrate each time (though perhaps longer would have been prudent, since I still saw low frequency drift up to 4 MHz during the 20 min measurement time). No improvement to the estimate on Teraxion laser, and if anything the West laser was even more noisy relative to Rio E and Teraxion, across a wider frequency band (almost the entire band from 30 mHz - 90 Hz). The foam and turning off the HEPA FFU did reduce the noise below 1 Hz, especially for the Rio E x Teraxion beat note. Figure 1 uses maximal averaging for every 2-fold frequency increase, on 20 minutes of data taken today (attachment 2 reproduces the relevant figure from last week's data, without the Marconi reference for better viewing).
Attachment 3 is the updated Teraxion noise estimate.
Update: attachment 4 is a comparison of the frequency noise (uncalibrated) for three different configurations: after taping down the fibers, after adding foam, and after turning off the HEPA blowers. I'm not sure why turning off the HEPA blowers increased the noise, maybe should have let the system settle longer? Despite the overall higher noise floor with HEPA blowers off, several peaks between 10 and 100 Hz were reduced.
We found a heavily pitted cantilever from Zach's early fab runs, and mounted it in his clamp using the alignment jig and pins. We tried optically contacting one of the mirrors to the cantilever, but some combination of surface oxidation and roughness (or just inexperience) prevented us from making a bond. The surface of the cantilever had obvious defects, so we weren't very hopeful. We're seeing what Koji suggests for glue.
We're starting to make new cantilevers this week. Here is our process, largely drawn from Zach's thesis, based on the process from the Chao group (D1200849) and the standard techniques of hard mask etching.
Zach started with a 100 mm (4") undoped <100> Si wafer with 500 um thickness. He reports achieving similar mechanical Qs of the final cantilevers when starting with either SSP or DSP polished wafers, though we may want to investigate this further. To avoid spoiling too many large wafers, I'd like to start by processing our 2" x 280 um wafers that are leftover from the cryo Q experiment, and will inventory our larger wafers to determine what we should order. We also have some 3" wafers we can use to fabricate full-length (7 cm) cantilevers in smaller batches than had we used 4" wafer.
Zach's procedure calls for thinning the central region of the cantilever, which softens the suspension and improves isolation. I expect we'll want to evaluate our cantilevers before thinning, since the procedure to thin the central region is a bit tricky. When we do thin the cantilevers, we will not terminate the process at step 6 but instead continue with the following:
Materials we need to acquire are in bold
Our training status on the equipment is green for 'full user,' orange for 'supervised user,' or red for 'need initial training.
rather than the perfect solution to PSD estimates, how about using the code you already have and just change the binning a little more often than once per decade? i.e. stich together as you already did, but get more averaging in the noisy spots. Should be a very easy modification to the code.
I modified Aaron's DFD box: I replaced the SLP-1.9 (1.9 MHz low pass filter) with a 44 MHz low pass that we conveniently had in the lab.
The purpose of this is to measure the high frequency noise of some of our lasers to figure out if there's any difference between the RIO Planex and Teraxion NLL.
I did the hookup and everything looks good so far. So far I have measured the RIO E/W beat and the RIO E vs Terax beat.
Need to do some more data processing to get plots.
Some notes along the way:
Attaching beat note noise spectra showing the noise of the DFD (with no pre-amp, so probably Moku input noise), the RIO Planex (E & W lasers from the Cryocav table), and the Teraxion NLL. Looks like they're similar in HF noise.
Yesterday (August 25), I measured another 10 min at 488 Hz of Moku phasemeter data for the three pairs of beat notes. All three lasers had been on overnight, so there was no longer low frequency drift of the Teraxion laser. Rather than amplifying then picking off the beat note, I sent the RF output of the 1611 directly to the Moku's phasemeter input.
Today, I've been figuring out how to get more averages out of our data. One approach (the one used above) is a modification of Welch's method:
The above procedure sacrifices some frequency resolution at the higher frequencies in exchange for additional averaging. The tradeoff with resolution is necessary, because the window size determines not only the smallest resolvable frequency, but also the spacing of frequencies in the spectrum. For the spectra from the previous elog in this thread where N=10, the total measurement time is 10 minutes, and the sampling rate 488 Hz, there is evidently more noise higher in the frequency decade (7-9e^n) than lower (1-3e^n). More consistent averaging can be achieved by setting N=2, but at the expense of most of the high frequency resolution (only 33 frequency bins survive the procedure).
One workaround is to modify the procedure so the frequency binning is mostly set at the beginning, by the 'highest resolution' available. Then, perform Welch's method with as small a window as possible while still resolving the frequency bins. Care must be taken at high frequency: eventually, the 'FSR' of our Welch's method cannot resolve an f_0 difference in frequency into an integer change in the number of samples per segment. The spectrum can either be cut off at that frequency, or the procedure can continue while accepting nonstandard bin widths at high frequency.
Other workarounds are perhaps less desirable. One could accept nonuniform frequency binning, and simply compute Welch's method for every available choice of nperseg. This would maximize the number of averages in each bin, but especially at low frequency, there will be substantial correlation between adjacent frequency bins. Another workaround is to save the entire spectrum wherever evaluated, then combine the data later. One must again worry about correlations between the measurements: at high frequency, we would be combining coarse data with many averages with fine data with fewer averages.
Another approach entirely is to do something smarter than Welch's method. In our meeting today, Chris suggested I look into multitapering. Spectral estimates can reduce bias due to leakage by introducing a tapered window, at the cost of increased measurement variance. Welch's method heals the variance relative to standard tapering by overlapping the windowed segments, at the cost of some frequency resolution. Multitapering instead minimizes loss of information by increasing the number of degrees of freedom of the estimates. The Here are a few resources on the topic:
While checking out Percival and Walden, I stumbled across parametric methods for spectral estimation -- those where an early spectral estimate is used to refine the procedure and spectral estimate iteratively. Perhaps up the alley of some recent discussions at our group meeting.
Simply applying Gaussian error propagation is not quite right, because the PSD is exponential distributed (the ASD is Rayleigh distributed, see Evan's note T1500300). Each ASD is Rayleigh distributed with
For a large number of averages, the central limit theorem lets us estimate the mean of each PSD, , with normal distributed uncertainty and variance . Our three corner hat estimates of the ASD are based on the scaled, root mean-squared sum of three such PSD estimates, so for each frequency bin we can estimate the variance of the laser's ASD by
I'll use the final equation above, along with the number of averages, to estimate the uncertainty in each frequency bin of the final frequency noise ASD of the individual lasers. In particular, the filled region is .
[OK, I'm having a lot of trouble uploading pdfs to the elog this week, even with rasterizing. I've dropped these figures along with one set for the case of 'Welch's with no averaging' onto gaston under /home/controls/cryo_lab/Figures/3CH ]
I think these results warrant a more careful measurement, especially in the decade around 1 Hz. Also, the error bars are obviously way underestimated.
We uploaded the data from moku to dropbox, and pulled it to spirou via the web interface. Should give the workstations dropbox.
Later, I realized we could probably have done better by sending the output of the 1611 directly to the moku. Instead, we were amplifying +16 dB then using an RF coupler to pick off 1% for the moku.
After plugging back in all the cables, I was struggling to get the cavity locked again or seeing the transmitted beam on the monitor.
Today, following Aaron's suggestion, I removed the connection from the PDH mixer IF port to the oscilloscope and also the power splitter I had added earlier (now sending the IF output directly to the LB input only with the 10 dB attenuator). The cavity didn't lock right away but changing the gain to 6 and re-adjusting the input and output offset got the cavity locked again.
When we tried to get it on the monitor, we saw that the mode it locked to was a higher order mode misaligned in the vertical direction. We first aligned the beam vertically until we started seeing lower order modes and finally the TEM00 mode.
I adjusted the lenses a little to tweak the mode-matching after aligning the beam into the cavity for maximum transmission when locked. Calculating the mode-matching using the reflected beam when locked and unlocked shows that it is only 60% though.
The noise I was seeing last week on the ADC did not show up when driving the same channels directly with a function generator, only when buffering the function through the SR560. Wrapping the BNC several times around a ferrite toroid between the SR560 and ADC reduces the noise to close to the level of the ADC noise floor (there is a < 5 count pkpk sine wave cross-coupled into the adjacent channels for a ~2000 count pkpk sine wave on the channel of interest, but the signal carrying channel itself looks clean).
The SR560 that was overloading on battery last week is now also overloading on line power. I've swapped it with one of our functioning SR560.
Afterwards, I looked at the spectrum for the Rio E x Rio W laser beat note. Looked OK... there's some kind of filtering happening in my delay line though. If I watch the spectrum of the RF coupler's pickoff (1% between the amplifier and delay line box), the implied peak power entering the delay line box is ~ 5 dBm near any of the nulls (so ~113 MHz or 176 MHz), but over 15 dBm between the nulls. I hadn't noticed this behavior before (with the busted mixer and RF coupler before the amplifier).
I'm driving the input of the delay line box with 10 dBm from Marconi's RF output.
Ack!! stay out of the "Fix the SR560s" game Do not repair, do not send back.
Let's get a trustworthy measurement of the frequency noise ASAP.
I've borrowed one SR560 from CTN to get us by for now, but am wondering if I should order parts to make repairs, or simply send the units to SRS?
I replaced the mixer from my delay line box with a functioning ZFM-2-S+. I also replaced the lossy pink SMA cable (122 cm) with a slightly longer (150 cm) cable that I made from a spool of LCOM coax. I also replaced the shorter cables with solder soaked ones. The final build is in attachment 1.
I drove the delay line frequency discriminator with a swept sine from the Marconi, and observed a symmetric response in the IF output. Sweeping from 5 MHz to 80 MHz is in attachment 2, showing that the response is symmetric about zero from -107 to 108 mA.
I'll return later this afternoon for a more careful calibration and to take some data.
I'm calibrating the delay line discriminator by running a swept sine from the Marconi, and this time reading back on an acromag channel.
Attachment 3 is the first run (y axis in Volts), where the Marconi is ramping from 1 MHz to 240 Mhz, stepping by 10 kHz every 100 ms. There are some surprises. The output is linear until about 70 MHz, possibly because the mixer is only good down to 5 MHz. The upper half fringe is not symmetric to the lower half fringe, even comparing the curve only above 10 MHz.
I'm repeating the measurement with different sweep parameters: 1 MHz to 250 MHz, stepping 1 kHz every 50 ms. The steps will be somewhat faster than the Acromag's sampling rate, but those units dither so it should smooth out. I'll post some plots with the time axis converted to frequency when the data come in.
I'm taking this data today, here are the issues I've hit
We have 7x SR560 in need of repair in the cryo lab. Most of these (at least 5) are constantly overloading, which indicates the front end transistor and op amp need replacing. From SRS, these parts are $50 each, but I found I think the same parts on digikey (the FET transistor is LSK389B; the amp is OP37A) for about $10 each. There are at least 2 that need a new battery.
Rana noticed that my delay line is asymmetrical -- sweeping the frequency through a full cycle of the interferometer should reverse the sign of the output, but instead the output was biased negative.
Indeed, when I drive the mixer RF with 0.5 Vpp at 25 MHz from a function generator while supplying the LO with 7 dBm at 150 MHz from the Marconi (see attachment 1), my ZFM-3-S+ has an asymmetric waveform despite at most a few mV difference in the LO and RF DC levels (see the video in attachment 2). The same test with a different mixer (ZFM-1-S+) gives the balanced waveform in attachment 3.
I'll replace the mixer, and make a few modifications to the cables of my delay line box.
I'm repeating this measurement, but moving the photodiode and electronics to the cryo cavs table, so we don't send anything over a long, floppy fiber across the room. See attachment 1 for diagram.
I set up the measurement, but didn't take any data. What's the right way to find the IP address for these GPIB controllers? I ended up scanning with nmap, but suppose I should mess with prologix' netfinder tools to assign our controllers static IP addresses and just label them.
could the pumpdown plot be made so that the units are visible? Maybe use dataviewer or python?
We repeated steps 1-4 in elog 2789 and, with two people, managed to get the o-ring to stay in place while lowering.
During the vacuum testing, I'm continuing some lab maintenance, mostly cleaning up and labeling cables on the electronics racks.
The cantilevers cryostat vacuum line [edit: Aaron (May 2022) suspects I was pumping on the vacuum line only, not the chamber, as indicated in the previous log] has only reached 7 utorr (the reading on the gauge matches epics) after pumping overnight. I'm going to try manually actuating the gas ballast valve in case our roughing pump is manual-only.
After closing the gas ballast valve, the pressure drops below 7 utorr in under 30 minutes and is approaching 1 utorr. That's not the best I've seen from this pump, but should be good enough to continue diagnosing the cryostat.
To that end, I'm installing a new 2-270 Viton o-ring on the cantilever cryostat (needs name).
Facilities came to move out the cryo Q (central, rigid legs) optics table, and left us the legs. They also took the two desks from the lab. Lastly, they moved the PSOMA table to the center of the room and rotated it 90 degrees so its long axis runs EW. Afterwards, Shruti and I moved the PSOMA rack to the E end of the PSOMA table, directly across from the other two racks .
Attachment 1 is a screenshot of the temperature and particle count trends over the last 4 weeks. I need to figure out how to add axis labels on ndscope... the units for both temprature channels are F, relative humidity is %, and all particle count channels are log(counts). There is a factor of 10-1000 excess particle counts around the time of our lab rearrangement, mostly affecting the larger particle sizes. Humidity also experienced a mild increase. The AD590 temperature channel drops off because I turned off its power supply to accommodate our rearrangement, and haven't yet turned it on again. The daily and weekly particle count fluctuations are interesting, presumably dictated by lab access.
Today we are testing the pumping station while pumping on just some blanked off Ts and the vacuum gauge. Last time, we pumped on the vacuum hose leading to the closed Key valve at the cryostat, and observed the pressure level off near mtorr, before eventually reaching only several utorr. The turbo should really have no trouble getting to utorr pumping on just hose sections, so we'll try to observe some better pumping action today.