[Jeff, Daniel]

We cleaned the non-vacuum portion of the 10" to 4.5" CF Zero Length Reducer Flange on Laser Filter Cavity (LFC) Output Vacuum Cube using wipes with isopropanol and acetone to remove sticky tape residue. This worked pretty well. We could not remove the copper gasket for the 4.5" CF Flange by hand or with careful use of a flathead screwdriver. This gasket is likely made of annealed copper and has been in place for ~10 years. We covered the area in UHV foil. Some sources on the internet recommend locking pliers. 

Continuation of 12055.

I touched up the alignment of OFC1 1550 path until it looked good enough to lock. I locked the cavity with the laser in order to see a constant stream at the output. I then used this to align all of the transmission optics outlined in the previous post. This works great and everything is aligned. Here is an overhead shot of everything in action. Note the three cables coming from the flipper mirror. One is power, one is a physical button located on the side of the table that can be used to flip between the transmitted light on a camera/PD combo or chaining to the next cavity, and the third is a USB that can be hooked up to the computer but is not currently because i can't find a USB extension (this will also be cool because when we make guardians to control the cavities these can interface with python I believe). I really like this design where we can quickly flip between diagnostics and cavity chaining modes. Other minor note is the flip up mirror path being the diagnostic over the connection to the fiber should prevent us from having to make minor adjustments into the fiber when changing modes, but this MAY be required if the repeatability of the flip up mirror isn't good enough to center on the PD/camera each time. After clicking the button 15 ish times it seems like its more than enough, just something to watch for.

Saved some screenshots of some quick cavity scans. This is showing the REFL and TRANS bd of the 0,0 mode. Note that it does not look like the cavity is close to critically coupled (we have never seen this cavity critically coupled even in B102). And then this shows the difference in the 0,0 mode and the rest of the higher order modes leaking through. This may be able to be improved via finer alignment and adjusting mode matching but is more than enough for the science we want to do now.

I think I want to retake some ringdown measurements to see how dirty the cavities became after the move and align the output into the fiber that connects to OFC2. Pending this and a quality piezo lock on both cavities we are ready for a carrier suppression measurement of both cavities.

[Ian, Daniel]

We removed the 10" to 2.75" CF Zero Length Reducer Flange from the Laser Filter Cavity (LFC) Output Vacuum Cube and put it on the "output port" of the Demonstrator Interferometer Vacuum Cube. We cleaned the vacuum and surrounding areas near this output port on the cube before placing the flange.

We removed a 4.5" CF Window (presumably AR coated for 1064 nm light), wrapped it in a dried out clean room wipe that was previously wetted in isopropanol for UHV, wrapped that in foil, put it in an anti-static bag, labeled, and put it away.

We removed the 10" to 4.5" CF Zero Length Reducer Flange now with UHV foil where the window was previously attached, cleaned the flange, and placed it on the newly opened spot on the LFC Vacuum Cube.

I then tightened the bolts on these two 10" CF Zero Length Reducer Flanges to 34 Nm.

I hope to finish work on the LFC Output Vacuum Cube tomorrow, which involves cleaning the 10" to 4.5" CF Zero Length Reducer Flange outside from tape residue, placing a 4.5" blank flange, and adding the custom 10" to 8" nipple.

[Torrey, Jeff]

Today we continued the re-alignment of output filter cavity one. The 775 path is now aligned well enough to lock, and the 1550 path is aligned well enough to see flashes in transmission but still needs input alignment tuning to suppress the higher-order modes.

As discussed in [12053], we decided to realign the first output filter cavity in order to avoid the 775 beam clipping the piezo mount on transmission. It was clear that the piezo controlled mirror was a bit crooked (pitched down), and this required a wonky input alignment which led to clipping. We dissasembled and reassembled the piezo optic assembly and found that we could change the tilt of the mirror significantly by changing the compression of the viton by a very small amount, from which we conclude that the viton is being scrunched and we had perhaps been over-tightening it. We also experimented with a few rotations of the assembly, and the configuration we settled on was a 90 degree rotation from where we started. After this the mirror was noticably less wonky, and we proceeded with realigning the cavity, doing the 775 path first as it is more geometrically constrained by exiting through the piezo.

We performed a coarse alignement by removing the 775 input mirror and then using the cavity mirrors to postion the round-trip beam to be coincident with the incoming beam at approximately the location of the input mirror. We then used an iris to mark the round trip beam after it leaves the cavity, and after replacing the input mirror we tuned it such that the prompt reflection hit the same iris. This worked well enough to see a second pass beam in transmission, from which we could do further alignment.

We tuned up the cavity and input alignment to maximize power in the 775 0,0 mode (see attatched scan on the photodiode).

Next we roughly aligned the 1550 in an analogus way by removing the input mirror, etc. We are now seeing lots of modes flashing during a scan, and will continue to tune up the input alignment to eliminate the higher order modes. We also started setup for a fiber coupler to send the light to the second filter cavity, using a flipper mirror to toggle between diagnostic (camera and PD) and fiber coupler. (see attatched diagram)

[Jeff, Daniel]

I tightened the 10" to 4.5" zero length reducer flange on top of the vacuum cube.

We added the CF0600S2 6" to two 2.75" CF Multiport Reducer to the 10" to 6" CF Zero Length Reducer. We did the usual method of placing the flange using some top screws to hold it, adding 2 bottom screws, remove the top screws, tilting the flange up, dropping in the gasket, and then adding the remaining screws. Only a few screws were accessible with our torque wrenches, so I tightened the screws by hand in a star pattern (going every 5 screws with 16 total). We bought this part (as part of a set) so that the torque wrenches can reach all the screws.

We took off a plastic cover and cleaned the 10" to 4.5" reducer flange, which was covered with duct tape (and duct tape residue) outside where the 4.5" flange goes (and outside the vacuum). 

We added a 4.5" CF blank flange to this 10" to 4.5" reducer flange on top of the vacuum cube. Because there were only 8 screws, I started at a lower torque, 6 N m, did a star pattern every 3 screws so that every screw was tightened twice, and incremented the torque by 1 N m until the torque was 12 N m, then I incremented the torque by 3 N m until 34 N m when the flange seemed tight enough (stainless steel contact around ~70% of the surface).

We added a 2.75" CF blank flange to the CF0600S2 (where the laser view port will not go) and attached it with the plate nuts. I lightly torqued by hand and used the torque wrench starting at its minimum value of 6 N m. I tightened every 2-3 bolts until the torque wrench did not tighten the bolts further, then incremented the torque by 1 N m until 14 N m. For the laser viewport, we will use the annealed copper gasket, which will require less torque to tighten.

Next, we will swap out the 10" to 2.75" reducer flange for a 10" to 4.5" reducer flange because the 10" to 2.75" flange is needed for the demonstrator IFO.

[Jeff, Torrey]

Over the last week I have been struggling to get OFC1 realigned. The 1550 path can be locked on a very clean 0,0 mode. The 775 light on the other hand caused a few issues, the main problem being it was resonant but badly clipping on the piezo as it exited the cavity. The 775 light looked like it was hitting the piezo mirror at the top left of the mirror. We then locked the 1550 light and used the SWIR camera to look at the 775 output coupler. This showed alot of scattered light, but no solid beam. After breaking the cavity this scatter disappeared completely, suggesting the resonant path in the cavity hits the piezo upon exiting. We took off the 1550 input coupler to confirm this and it was hitting the same top left portion of the piezo mirror. 

With this information we have decided to adjust the inter cavity mirror alignment. The piezo mirror has a significant amount of down tilt since it was first contacted and re-set. Our plan tomorrow is to align the 775 input, but first take apart the piezo mirror and reassemble so that it sets in a different alignment. 

I've installed a few of the fiber holders that Ian suggested we purchase. They seem pretty decent. A few of the fibers at the inputs got misaligned when feeding them into the holders and I corrected for this. The main drawback was that they take up a large amount of space compared to just looping them. This may be a problem as we become more progressed in the experiment.

Previously [11987] when attaching the Wen drill press to the table, the base of the drill press broke and sent the drill press falling to the ground off the table. I contacted Wen and they sent me a replacement base for free. I have now attached the replacement base to the drill press and stood the drill press upright as seen in the attached image (Drill_Press_Upright.jpeg). I turned it on to make sure it worked and it seems to be working fine. the chuck fell out of the drill press when it fell and will need to be reseated. Next I will try to attach it to the table again.

I tightened the screws 10" to 6" zero length reducer flange to 34 Nm in which most of the flange contacted the vacuum cube and the copper gasket could no longer be seen. It took 40 Nm to tighten the screws on the 10" diameter electronics flange such that there was this contact. This is a flange from the Holometer, so maybe they require this higher torque spec. Alternatively, this could be a coincidence. There is a small notch milled out of this flange, but it shouldn't make a difference.

I have moved the Intel NUC named Gouy from B102 to the control room in B111B since it does not see use in B111B and it is already set up with a CDS virtual machine.

I connected this machine via ethernet to the wall, and initially it said 'unidentified network' and 'no internet.' After setting the DNS address to our internal DNS, it successfully connects to the internet, but it says the network is called 'pLANck 2', what is this network? It appears to be on the subnet, it recieved a '248' IPv4 address, and on windows it is able to resolve names stored on our DNS.

In order to boot the VM I had to change the network adapter settings. Inside the VM it is connected to the internet, but it is not able to resolve names from the local DNS.

cat /etc/resolv.conf
# Generated by NetworkManager
search caltech.edu
nameserver 131.215.254.100
nameserver 192.168.248.15

It has our local DNS listed as an alternate, but it does not seem to ask our DNS if the Caltech DNS says 'not a thing.'

https://superuser.com/questions/681357/alternate-dns-server-not-resolving-after-first-one-fails-to-resolve

https://unix.stackexchange.com/questions/577222/alternative-dns-server-cannot-resolve

[Torrey, JC]

JC and I picked the lock of the locked lista cabinet in the cryo lab and got the IR viewer. I used it to observe the power distribution center. I did not observe any significant amounts of scattered 1550nm light. There was a fair amount visible on the SHG sled where irises are being used to block the 0th order beam but this is 775. The amount of power there is known to be low though, I don't think this is an issue. 

In short, I'm not sure we need one of these viewers. I could not observe any scatter where I expected there to be some. It could easily see 775 nm, as expected as they are several orders of magnitude more sensative at this wavelength. One caveat potentially is this looks like a much older model than what is currently being sold, maybe the technology has improved since this older version.

[Alex, Sander, Daniel]
 

The Thorlabs parts to make a cart with a breadboard top did not fit together. In addition, the POC001 cart could shake pretty easily with the included wheels. I therefore designed some custom plates, one set to attach new wheels from McMaster and another set to interface from the PWA075 to the B2436G Breadboard. To attached the plate to the cart for the wheels, I used Stainless Steel 15 mm long, button head M6 screws. To attached the McMaster Wheels to the plate, I used 3/4" long, 5/16-24 screws with 5/8" OD washers.

To attach the plate to the breadboard, I used a 3/4" long 1/4-20 button head screw with a lock washer (similar to this product) to prevent the plate from rotating (parts held by a single screw without a lock washer tend to rotate). To hold the PWA075 to the breadboard, I used some Thorlabs stainless steel 5/8" long, 1/4-20 screws with some washers.

To attach the PWA075 to the cart, I used a 55 mm long M12 set screw that I screwed into the PWA075 since that was a blind hole. We used 8 rotations to put the set screw into the cart. I used nuts and washers to prevent PWA075 and the cart from spinning on the set screw.

When possible, I used alloy steel fasteners for its strength. The parts from McMaster were fairly greasy, so I washed them in simple green in DI water (1:30 ratio), sonicated for 10 minutes, and washed in DI water.

I made the plates out of Aluminum 6061. Attached are the part files, STEP files, ProtoMax Waterjet machining files, and drawings for making the parts in the mill. I had to machine 2 plates that interface with the breadboard as drawn and 2 parts with the opposite face milled because the parts need the be flipped accross the axes transverse to the breadboard face. The plates were greasy, so I washed them in simple green in DI water (1:30 ratio), sonicated for 3 minutes, and washed in DI water.

We assembled everything together, and everything fits. See attached photo.

Jeff and I met with the Moku CEO and several employees today. Meeting went well. One take away I wanted to make public knowledge is the different capabilities of the computer app vs the ipad app. Attached are some example screenshots. moku1.png can take a spectrum and a transfer function of the same signal at the same time, something I have not been doing up until this point. moku2.png as a starter point for when we are chaining cavities together in the future. 

[Ian, Torrey]

We noticed that it is very humid in the lab today. It is also much warmer than usual. The attached plots for the RbQ lab (A) and the GQuEST lab (B) show a spike in humidity between the 8th and 9th of December from about 20% relitive humidity to about 50% relitive humidity. The water does not seem to be condensing.

I have followed this guide to set up software on the CyMAC front-end machine running bullseye (Debian 11), also known as babbage.

Notes/Questions:

• What is the SITE variable used for? I assume this irrelevant for us since we are not building a gravitational wave detector that will share data with the other LIGO sites. I set it to the default 'TST'
• Another prompt asks about 'Kerberos Authentication', and asks for a hostname. I set it to 'caltech.edu' which appeared automatically in the box.
• I left all the remaining fields related to Kerberos blank, including hostnames of Kerberos servers.
• I left blank the location to store .mdl files, as it said it could be set later. How do we want to organize the filesystems of the CyMAC machines?
• I ran 'rtcds status' and got

Kernel Module Status
mbuf               ***NOT LOADED**
gpstime            ***NOT LOADED**
rts_cpu_isolator   ***NOT LOADED**
rts_logger         ***NOT LOADED**
Kernel Module Status = SOME MISSING

System Status
system         epics          module         awgtpman
------------------------------------------------------------
Systems Status =       ALL ACTIVE

Streamer Status
rts-local_dc       OFF
rts-transport@cps_xmit OFF
Streamer Status =      DEGRADED

Overall Status =       DEGRADED

• why might the 'Kernel Modules' not be loaded? 

[Jeff, Daniel]

I moved the "output" Laser Filter Cavity Cube closer to the edge of the table so that we could work on it more easily. Jeff and I took off the 10" CF flange with an angled 2.75" CF flange off the cube, wrapped it in foil, and put it away with the other holometer parts. I didn't like this flange as it stuck out past the custom aluminum base and didn't seem very useful. In its place, we moved the 10" CF flange with two 15 pin D-Subs from the top to this location (this flange will be facing South when the cube is placed back). We moved a 10" to 4.5" zero length reducing flange from the side to the top. In that place, we added a new 10" to 6" zero length reducer. I believe we match up the hole pattern so that the 6" to 2x2.75" reducer will have the windows 5" above the cube bottom as designed.

I cleaned parts of the cube and the flanges with the Wipes rated for cleaning UHV, but the surfaces were pretty clean to begin with.

We only tightened the screws so that the washers couldn't move. I still need to tighten the screws to compress the copper gasket.

[Sander, Daniel]

We routed a USB Extension Cable from the vacuum oven in B110 through B111D to the B111B control room. The cable trays above the mobile clean rooms in B111B were very hard to access. Maybe we should get some sort of cheap grabber tool from Amazon.

The cables make it to the computers in B111B but just barely. A ~6 ft USB extension cable would ensure there is plenty of slack in the system.

The computer recognized that something was plugged in, so it seems the extension cables work. Next week, I will finish setting up the vacuum oven software that I downloaded. I previously got this software to work on my laptop.