[Lee, Jeff, Daniel]

We plugged in the ion pump controller to a wall outlet and read the current from the wall outlet to the ion pump controller as 0.03 A. The voltage from the BNC on the front of the controller was 9 V.

The pressure was 2.5E-7. We hit enable and the on button around 2:40 PM. The pressure jumped to 1E-4 Torr and then quickly dropped. The BNC voltage dropped to 2 V before slowly rising to around 3.1 V. The controller drew 0.7-0.9 A. There was no need to repeatedly hit the on button; the ion pump stayed on and did not trip.

At 4:28, the pressure was 1.9E-7 Torr. I shut the gave vale to the turbo pump and the pressure rose to 3.3E-7 Torr. I turned off the turbo pump and the pressure again jumped to 3.8E-7 Torr. By 4:47 PM, the pressure was 4.7E-7 Torr. No turbo pump speed was given and it sounded quiet, so I turned off the scroll at 4:49 PM. The BNC voltage was 3.95 V.

By 4:54 PM, the pressure was 5.1E-7 Torr and the BNC voltage was 3.93 V.

After turning the turbo off, I believe I enabled a setting so that the turbo speed will be displayed as its spinning down. The controller is unfortunately hard to use.

I will monitor the pressure in the coming days. If the pressure approaches 1E-4 Torr, I am going to turn off the ion pump to protect it.

Going forward, it might be nice to add a valve to the ion pump so that all systems can be vented nicely. Maybe one of these up-to-air valves (cheaper than on Lesker) and a CF Tee.

I have successfully gotten the Google doorbell to work via python

However, Google restricts your ability to pull media from the doorbell api if you are not using their home app, so when I see the video clips, it returns a 401 error, which indicates that I am not accessing via the home app

There is a work around, but this will take some time to test and may not be a priority: https://medium.com/@tamirmayer/google-nest-camera-internal-api-fdf9dc3ce167

My code currently uses the Google API and a refresh token to try to access the image preview (we will see if it works after trying the thing above)

The doorbell now uses pub/sub to post events, and my python code allows the user to subscribe to the event log and pull event threads when available. The events for a doorbell press (sdm.devices.events.DoorbellChime.Chime) require no other user interaction and can trigger the doorbell notification via Mattermost once I work with Ian to set this up. The result is:
 

Listening for messages on projects/lab-doorbellapi/subscriptions/doorbell-subscription-v1...

Received message: b'{\n  "eventId": "b3326428-aceb-461a-ab2a-25ab3f8c06a6",\n  "timestamp": "2025-04-12T02:00:15.064852Z",\n  "resourceUpdate": {\n    "name": "<redacted>,\n    "events": {\n      "sdm.devices.events.DoorbellChime.Chime": {\n        "eventSessionId": "1740690860",\n        "eventId": "n:1"\n      }\n    }\n  },\n  "userId": "<redacted>",\n  "eventThreadId": "6855bbc9-f284-4582-a679-b875450c9ec2",\n  "resourceGroup": ["<redacted>"],\n  "eventThreadState": "STARTED"\n}'
Processing event for eventThreadId: 6855bbc9-f284-4582-a679-b875450c9ec2
Doorbell was pressed! 
Received message: b'{\n  "eventId": "b3326428-aceb-461a-ab2a-25ab3f8c06a6",\n  "timestamp": "2025-04-12T02:00:15.064852Z",\n  "resourceUpdate": {\n    "name": "<redacted>",\n    "events": {\n      "sdm.devices.events.DoorbellChime.Chime": {\n        "eventSessionId": "1740690860",\n        "eventId": "n:1"\n      }\n    }\n  },\n  "userId": "<redacted>",\n  "eventThreadId": "6855bbc9-f284-4582-a679-b875450c9ec2",\n  "resourceGroup": ["<redacted>"],\n  "eventThreadState": "ENDED"\n}'
Processing event for eventThreadId: 6855bbc9-f284-4582-a679-b875450c9ec2
 

The print statement "Doorbell was pressed!" occurs when a new event is processed and its eventThreadState is "STARTED" 

This will be changed to a Mattermost post via webhook in the future

Access to the developer settings for the device console, pub/sub, API, and Google Cloud developer settings are all set up for gquestlab@gmail.com 

Please don't make any changes to the Lab-DoorbellAPI project without contacting me first.

[Ian, Jeff, Daniel]

We connected the IDP 7 Scroll Pump for the Laser Filter Cavity (LFC) to the same power system as the SNSPD Dewer pumps. We connected the Turbo and Turbo Controller to a power strip on the top shelf in the mobile clean rooms. The pressure gauge, after warming up for a few minutes, read "pressure too high". This is definitely a measurement error, but I think it's fine. I plugged in the USB-B extension cords from amazon and they successfully power the fan that cools the turbo pump.

We turned on the scroll pump at 4:24 PM, and after ~30 seconds it made a secondary "turning on" mechanical sound and the pressure gauge started to read actual values instead of "pressure too high". After 20 minutes, the pressure was 0.1 Torr. We plugged in the interlock bypass D-Sub and turned on the turbo. After 2 minutes (and 1 vertical bar on the turbo controller), the pressure was 3*10^-4, after 12 minutes, the pressure was 8.7*10^-6 Torr. The 3 vertical bars then revealed the turbo speed, 1167 Hz. The final power was 12 W after peaking around 90 W while ramping up. I left after 15 total minutes of the turbo being on with a pressure of 6.9*10^-6 Torr.

I connected the ~2 ft long KF 16 to KF 25 tubing from the Agilent TwisTorr 74 FS Turbo Pump to a 4 ft long KF 25 tubing which was connected to the Agilent IDP 7 scroll pump. I used the included filter inside the o-ring going to the scroll pump.

Now that the vacuum system is fully connected, I opened up the gate valves to the turbo pump and the ion pump. I also connected the power cord to the scroll pump but didn't plug it into a wall. I flipped a red switch so that the scroll pump is expecting 110-125 V instead of 230 V.

[Alex, Daniel]

Alex finished the assembly of the SNSPD Dewer and vacuum parts. We then pumped out the vacuum system with the IDP 7 scroll pump to 8.7*10^-2 Torr (there is some systematic uncertainty with the pressure measurements with the pressure gauge and the magnet location). We enabled soft start on the turbo pump rack controller (see the wiki for how to) and plugged in the pump (while the rack controller wasn't powered). The soft start, which slowly ramps up the pump speed which is good when the pump has been off for over 3 months) remained on after this power cycle. We plugged in the included female 9 pin D-Sub with pins 3 and 8 shorted into P1 to bypass the interlock. We could then turn on the turbo pump. The pressure immediately dropped to ~10^-3 Torr and the turbo increased up its speed (not displayed by the controller) over ~15 minutes. The power draw by the turbo started at 10 W and peaked around 90 W. After these 15 minutes, the pressure was 1.6*10^-4 Torr and the rotation speed was 1167 Hz, the nominal value. The power drawn was 16 W.

I am struggling to align the 1550 input on OFC3. We previously solved the problem of have a stable enough test light to align with, but a new problem has arose where the output of OFC2 1550 light is tilted down at a fairly steep angle, enough that the beam is barely on the first intercavity mirror. The mirror mounts do not have enough range to correct for the translation required for the new input beam. After lunch I will attempt some kind of periscoping mirror solution in order to raise the input beam 1-2 inches between the two intercavity mirrors. I believe we have the parts required but am not sure. May need to order a few things.

This problem could easily be solved by touching the intracavity mirrors, but because they are already very well aligned for the 775 path (confirmed via cavity piezo scan), I think it is best to make a periscoping mirror solution work for 1550 light.

I met with Tim this morning to property tag the new fiber eom that is in use. Two things to note:
1) We decided not to physically tag it. He took a picture of it and is going to use a "no tag" classification in his write up. Expect for them to look for this in two years.

2) This order had 3 fiber EOM (2 phase 1 amp) and he didn't want to tag the other two or just didn't know they existed. Just making a note so people are aware only 1 is technically property tagged.

I unpackaged all of the items for the Infleqtion Rb Cold PICAS. I couldn't initially figure out how to attach the "DN40CF (2.75" flange) to DN16CF (1.33" flange) adapter with mounting points 2D+ MOT coil assembly (CAAMPA)" to the laser and coil parts. I tried unscrewing what I think are M4 screws (they took an M3 hex driver), but I could only get 1 out and the other 3 spun without loosening anything or changing the height of anything. I belive these screws go into posts and the laser and coil parts pull away from each other. I am meeting with Hannah Manetsch, who assembled a very similar version of this project but with Cesium, this upcoming morning to figure this out.

I also inspected the glass cell with a 1.33" flange that is our Rb source. It looks good.

I added a blank 4.5 in CF Flange to the top of the cube, and 2x25 pin D-Sub 4.5 in CF Flange to the side of the cube, and 4.5 in to 2.75 in reducing 4 way cross to the top of the prism for the RbQ MOT Chamber. I tried the molded viton gasket, but it seemed a bit too large. The unmolded viton gasket, even for the "vertical" flange orientation where the gasket doesn't sit nicely, were fine. I treated the gasket like a copper gasket by putting it on the flange and quickly bringing the flange and gasket toward the chamber.\

I tightened all the bolts with a small wrench, a small socket wrench, and finally to 50 in lb (5.7 Nm). We need a torque wrench that can go to lower values.

[Jeff, Ian, Daniel]

Ian and I machined out 1/4-20 holes in the Thorlabs MB1824 aluminum breadboard into 0.266" diameter through holes for 1/4 screws so that we can put screws into the blind holes on the Ideal Vacuum base plates. We used the drill press where we could and a drill otherwise for holes too close to the center of the breadboard. I accidentally flipped the orientation of the holes, so I am currently building the mirror image of my SolidWorks design. I don't think this will be an issue. If it is, we can machine more holes into the MB1824 breadboard.

Jeff and I cleaned with isopropanol wipes and attached the two ideal vacuum "cubes" following these steps. We put the "high pressure side" away from the MOT because I figured it would be easiest to screw in parts there. The cubes were flush together, but the cube was oriented a bit clockwise when looking down from the cube to the prism.

We then attached the sides to the cube/prism assembly. For the 6"x6" plate to 2" diameter window, we cleaned and dropped in one viton gasket but couldn't fit a second gasket between the window (which we cleaned with puffs of clean air) and the "screw", partially because the mirror is too thick, partially because the viton is too wide when it's on the "screw". I tightened the "screw" finger tight. This can't be good for the window, but the edge of the window can't be used anyway. The "screw" sticks out approximately 0.75-0.76" for all 3 we assembled.

For tightening the plates to the cube/prism assembly, I did a few turns by finger to get the screws in, then used a screw driver until I encountered some resistance, then did a bit more on the screwdriver, before finishing with the torque wrench at 50 in lbs (5.7 Nm) and then 70 in lbs (7.9 Nm) twice, following these instructions. A torque wrench that can be set lower (like this) and corresponding hex bit socket set, if needed, (like this) could be useful. I used a lot of isopropanol wipes since the cleanliness of the parts is suspect. Hopefully everything is sufficiently clean.

Lastly, Jeff and I mounted the prism/cube assembly to the breadboard with some 3/4" long 1/4-20 screws and washers.

Next, I am going to attach parts to the CF flanges on the ideal vacuum plates. I am not sure which viton gasket I should use. (1 or 2)

I cleaned the 10" to 4.5" CF Zero Length Reducer Flange and the two 10" to 2.75" CF Zero Length Reducer Flanges on the Demonstrator IFO Vacuum Cube from duct tape and its residue with wipes, acetone, and a bit of isopropanol. The 10" to 4.5" CF Zero Length Reducer Flange has some grime that I partially removed. I tightened the bolts on the 4.5" CF Flange to 27 Nm which seemed to give a steel to steel contact such that the copper gasket wasn't visible. I tightened the bolts on the 2.75" CF Flanges to 18.6 - 19 Nm which gave a steel to steel contact. Next steps are to install the 10" to 6" Zero Length Reducer Flange, the 6" 6 way cross and all of the vacuum parts off of it.

I set up the piezo control for the third filter cavity. This is to allow a way to scan cavity 3 while keeping OFC2 laser locked on 1550. The scan works and after tuning the controller a little it provide one of the best piezo locks we've seen in these cavities. Note that this new cavity is with the new mirror mounts and the nac2125 from CTS piezo. More analysis is needed but this piezo seems promising. For now I want to align the OFC3 1550 path, which I now should have all the tools for. Will come back to this new piezo analysis later.

Continuation of this work.  Yesterday, Jeff and I set up a way to have constant light exiting OFC2 to align the 1550 path for OFC3. I placed lenses according to the mode matching solution in the previous post and profiled the light coming out. Plugging in these numbers yields less than 1% mode mismatch for cavity 3. I will attempt to align 1550 light into the cavity now.

I tried to power the ThorLabs ULN15TK with a UX4 EXCELSYS power supply. It seems as though we have the XgF/Xg8 powerMods installed, each with a max current draw of 3A, so I wired two modules together in parallel to surpass the 4A asked for by the laser.

Plugging in the laser results in a green power light and no status light, but the emission light does not turn on after hitting enable.

We should perhaps buy another DS12 to power the ULN15TK, but I remain confused as to why the other power supply did not work. It is providing the correct voltage, but perhaps wiring two modules in parallel does not increase the current supply as I expect?

The state of the lasers is that both lasers and the amplifier are off. The signs remain on.

[Alex, Daniel]

I got two aluminum KF50 Centering Rings from Nick Hutzler's group and machined them in the lathe to remove one of their lips so that they can hold a PCB that interfaces with the inside and outside of the Dewer. I clamped them on the outside lip with a 6 jaw chuck. I used a moderate amount of clamping force (~90 degrees of rotation with the chuck key) to hold them. I slowly increased the spindle speed to ensure the rings wouldn't fly off; I used a final spindle speed of 1700 rpm, which is pretty standard for aluminum and a carbide tool. I kept the x-axis of the lathe at around 1.93" and slowly moved the tool in the z-axis to remove the lip. I needed to take off 0.080" of material. For the last few thousands of an inch, I moved the tool in the x-axis instead of the z-axis. I then used a deburring tool by hand to remove any burrs from machining. I could not see any markings from the clamping jaws.

Tomorrow, I will clean and start to bake out these centering rings.