[Ian, Sander, Daniel]

I designed some guardrails that go 3" above the M6 tapped holes of the shelf above the optics tables in B111B. I used 1/8" thick, 1/2" wide aluminum bar. There were 3 piece types (all holes centered along the 1/2" width):

Vertical bar: 4" long; 0.24" (C drill) diameter hole 0.5" from bottom; 0.27" (H drill) diameter hole 3.5" from bottom

Long horizontal bar: 36" long; 0.27" diameter holes at 0.78", 1.5", and 3 more holes with a regular 250 mm spacing beyond 1.5" to match the shelf hole pattern.

Short horizontal connection bar: 19" long; 0.27" diameter holes 0.375" from side and 2 more holes with 9.125" spacing

This short horizontal connection bar is designed to be replaced by a patch panel if desired.

Ian helped me drill some holes and Sander helped me install the connection bars.

We were not able to install 1 long and 1 short horizontal bar because the seeder breadboard overhangs the shelf by ~1". 

See attached photos.

[Sander, Jeff]

First we installed the H1A:PCIe Gen 2x1 Host Adapter into the front-end machine named Babbage.

lspci | grep "PCI bridge"

the above command shows devices connected via the host adapter.

We next installed the advantech ADC, DAC, and digitizer card (also a DAC) into the 4U expansion chassis. We can then see new devices on running lspci, but the digitizer card is still invisible, and we believe this is due to a missing driver.

Boot sequence:

power supply to expansion chassis

power on backplane board

power on front-end machine

I've set up 3 ant traps at the base of the optics tables in B111B. This post is mainly a reminder to replace/remove them in 3 months. The extra traps chub gave me are located above the red tool chest in B111D.

I checked the dimensions, the custom SM05 threads, and the fits with the other parts for the custom 3D MOT holder from Hubs. We have previous issues with bad surface quality and bad SM05 threads. As far as I can tell, everything is up to spec (for the one part I checked). Everything fits together and the SM05 threads are good. I marked in the box which part I checked. The surface quality isn't amazing and could out gas a bit more than we would like, but I don't think we can get around it.

[Jeff, Daniel]
 

We cleaned the two recently machined 10" CF Flanges with the custom tapped pattern. We did a bath of 30:1 simple green + DI water and the sonicated it directly in the sonicator in DI and then isopropanol. We noticed a bit of a "scum" on the flanges. We aren't sure of the cause. We then put the flanges in the vacuum oven to bake out at 200°C for 48 hours.

I printed and posted these two signs to make using the drill press easier. The first is a chart that converts from diameter to drill size name. It also has what drill should be used to make tapped holes. The second sheet provides the recommended drill speed as a function of drill diameter for drilling into aluminum. I got this from the Caltech MCE Machine Shop.

After getting the 10" CF Flange from the CCE Machine Shop, I added the 2 grooves needed to vent the hole that had the broken tap. The hole that had the broken tap did not need to be rethreaded. I used a spindle speed of 3000 rpm and a feed rate of 2" per minute. I went approximately 0.015" deep like the other grooves, but it's hard to say exactly the depth with the mill I used.

[Torrey, Daniel]

I designed a part to hold this flexible metal conduit around the laser amplifier input and output fibers. It works by using a set screw to clamp on the conduit. It is made of 2 parts so that it can easily be interchanged around the fiber. It mounts on the #10-32 screws on the 19" rack. 

After making these parts with the water jet and mill, I cleaned them.

Torrey and I then put the amplifier output through the tube using gravity. We protected the amplifier output using kapton tape.

Attached are SolidWorks Part Files, STEP files, WaterJet files, Drawings, and photos.

I attached an aluminum wire screen to the back of the AOM Amplifier 2U Chassis to (hopefully) prevent RF radiation from going around the lab but allowing for airflow to cool the amplifiers. I needed to cut the chicken wire to allow for the #4-40 screws to go thorugh. I also needed some 0.281" outer diameter washers to hold the chicken wire since the screw head diameter went through the (cut) chicken wire. The wavelength of 200 MHz light is 1.5 m, so this chicken wire should be sufficiently dense to contain most of the RF output. See attached photo.

I've been trying to make the wireless work well with the new subnet. Looks like the only way to do it is to switch the router into access point mode, where it just acts like a switch that included connections over wireless. It forwards the outer network's DHCP and other broadcasts.

The downside of this mode is that it can't use the VPN feature on the router, but we will eventually get that more natively on our network (using CIT credentials) once they set it up.

I tried several things before settling on AP mode. You can attempt to connect the WAN through one of the LAN ports and disable DHCP + NAT, but I never could get it to pass the outer network DHCP, so I gave up. In the process, I accidentally put an access restriction on the user interface and locked myself (or anyone..) out of the machine. Thus I reset it. We lost the existing IP/MAC assignments, but we shouldn't need them anymore on the new network.

The VPN is also down now.

The IPADs on the wireless should see wired Mokus and should see devices from the other lab. You can move the router into B111 if needed, and we should buy more access points.

I cut a corner out of a tile (of unknown material) that goes in the mobile clean room using a bandsaw. This allows the patch panel cables to go from the optics table to the rack. The tile was easy to cut and didn't generate any dust.

The 16 port patch panel has been installed in the rack and on the optics table. We have made 16, 24ft cables to go between these two patch panels. These cables are completed and installed. During testing we found one shorted cable and replaced it while strung up. 3 cables have a small piece of orange tape on them as a warning that the pin that makes the connected may be recessed too far in the cable. All 16 have been tested and are functioning however. Currently they are simply labelled according to numbers on the patch panel. We may update this labeling system. At minimum we will add these numbers to the cabel themselves in case they are unplugged for whatever reason.

To add support for the lab DNS naming please follow this guide to add the DNS server (192.168.248.15) to your network adapter settings on Windows.

You should then be able to call devices by their set names on the lab network

After doing this I am now able to use the DNS name to call on my scope, ping it, and use within python frameworks

I machined and cleaned 3 more AOM mounts like in this post. I mounted one and put the others by the SHG sled.

I tapped the remaining 36 holes in the 10" CF Flange with a new tap. I raised the tapping depth by 0.0463" in case it was a chip in the bottom of the hole that broke the tap. I then added the groove pattern with the 1/8" end mill. The grooving would have hit the broken tap and possibly damaged the end mill, so I did not groove into it.

I will go to the chemistry machine shop tomorrow to see if I can use their sinker EDM to remove the tap. I will then retap or remake the hole, add the grooving pattern, then clean the two 10" CF Flanges.

We have two EVAL-ADH4702-1CPZ evaluation boards which are nominally configured as 21V/V gain noninverting amplifiers.

First I provided power with a Newport +-15V supply, connecting +, -, and GND on the power supply to VCC, VEE, and GND respectively, on the eval board.

This resulted in a gain of 12.3dB up to about 100kHz

Next, I used the on board power supply by connecting the DC jack to the +15V and GND of the power supply. I then connected the power supply half of the eval board to the opamp circuit by jumping VCC_TRANSFORMER
to VCC, VEE_ TRANSFORMER to VEE, and GND_TRANSFORMER to AGND, as described in the manual of the eval board in the section "Initial Power-Up." These jumper wires are white, turquiose, and yellow.

I performed a transfer function and a pulse delay measurement.