I finally got the scanning of 698. There was a single line in the code that should not have been there. And it was resetting all the data just before it was exported.

I adjusted the code to need a Scan Rate. I had some trouble with exporting, I got an OSError: [Errno 16] Device or resource busy: <filename> This was solved by not having the +'kHz' at the end of c1.filename, but it went away. After talking to grady, its openning/calling something too quickly.

the base Channel power is -65dBm and all images are base on 200mVpp Amp form SBench6

From the Spectrum Card w/o 30dB attenuator

100kHz Scan Rate

10kHz Scan Rate

1kHz Scan Rate

RF amp output, w/ 30dB attenuator, with swofl1 switch OFF

100kHz Scan Rate

10kHz Scan Rate

1kHz Scan Rate

RF amp output, w/ 30dB attenuator, with swofl1 switch ON

100kHz Scan Rate

10kHz Scan Rate

1. Open files folder tab

2. click This PC

3. Click the Computer Tab

4. Map Network Drive

5. In Folder name type the IP address of server

5.5. IP address is found in the same location where you connect to a network. Most likely called Unidentified Network. IPv4 is 192.168.2.2 in step 5, put \\192.168.2.1\<name-of-the-server>

<names-of-the-servers> = [common, strontium, strontiumwg]

I created a copy of Spectrum.py and ClockScanning.ipynb called Spectrum_Changable.py and Clock_Scanning_Changable.ipynb

I added in the self.view_freq_waveform() and self.view_waveform() plots so i can see how the frequency is changing over time. While viewing the Freq vs Time plot, i noticed the freq would linearly ramp from bottom to top and than go to zero. which could explain why i dont see any thing on the spectrum analyzer. I looked the Spectrum Wiki and notice the parameters for the plots were displayed. The Freq vs Time graph was periodic and that what should except to see when scanning the frequency. The Function in the code that was different was the self.set_freq_ramp().

Figures are stored in /media/strontium/engineering/Clock_Transistion/Scanning_Code_figures/

The self.set_freq_ramp() as 3 parameters that caused an issue, Duration1, Duration2, and Cycles.

PulseTime=180ms

The original was self.set_freq_ramp(duration1=PulseTime, duration2 = 0e-3, cycles=1), this gave:

self.set_freq_ramp(duration1=duration2=50us, cycles=100)

self.set_freq_ramp(duration1=duration2=50us, cycles=10)

self.set_freq_ramp(duration1=duration2=100us, cycles=100)

To clarify what i believe is happening:

The Cycles is the number of full sweeps (bottom to top to bottom) the function is generating.

Duration1 + Duration2 is the Period of each cycles. example, dur1=dur2=100us, cycles=100. the total periodic function will stop at Period*cycles= 20ms, this is shown in the graph.

Went to lab for a bit, to work on scanning code. I believe I have check all the settings in the spectrum analyzer, but still unable to do the scanning. Switched to looking at the code that produces the scanning Bin file.

I took Grady's advice to look for Python code that Paul used to create files for the spectrum. I looked at the code and noticed there was no way to export any code, but when i looked into Tyler's notebook, he talked about going through the AWG generation code located in media/stronitium/code/OFL/waveforms and here lies the code that creates the Scanning code called ClockScanning.ipynb.

With that said , I noticed that the Date Paul created the clock scanning code for tyler was the same date has the F=80_Sweep_1.0MHz.bin file was created and that the same name as was would have been exported from ClockScannig.ipynb this makes me believe that the .BIN file i have been using is correct and i Need to do further adjustments into the SBench to output the 1MHz scan at 80MHz.

learn frequency modulation

I learned a bit more on how to use SBench6. We have some Bin files in Strontium/eng…/Spectrum_awg. I used f=80_sweep_1_MHZ.bin ,but i unsure how to chagne anything. The demo Pro version only has 6 attempts left. I have used 3 attempts that just ended in crashes. Maybe we can get the pro version? I plan to stay in Base mode just in case.

Andor emailed back but didnt fully answer what chillers were best. They said, our current one is best for the iXon EMCCD, waiting for a response.

I spent the week grading and studying for my final on wednesday since its worth 60% of the grade.

I tried to get scan of the 698. WHile i havent yet, I have learned more about the Software. When i get an “No channels assigned to output” error, its most likely means the Digital output is enabled, you must disable all digital output.

I learned of the different functions i can use in SBench. They are located in C:/USers/paul/SB6_ApplicationsData/functions

Monday:

He looked at the tweezer intensity spots of a 10×10 array onto a camera via the Flipper mirror. He mentions the intensity is worse for lower arrays, and probably due to the phases. and with the 10×10 array, the intensities could be off due to a miss alignment

Tuesday:

Grady did made 698 Broadening Graphs and Talked a lot about tuning RF signals through the AOD, which a lot of it went over my head

Wednesday:

He work on the imaging paths, Put the 500mm lens back instead of the 750mm. We can now load atoms with uW per tweezer.

Thursday:

work on the imaging path some more.

Friday/satruday:

Grady found the 497 telescope was converging. and that there is an astigmatism, which was probably the cause of the elongagted tweezers. DUring the search for the antisigmatism grady found out the mirrors were clamped too hard. and that cause some of the affect but not all of it. He check all the mirrors. but the path wasnt aligned back.

Sunday:

Grady wants to redo the AOD to see if it gets better. but that for later.

I spent the day working on Sbench. I learned that how to apply different projects to different changes, or how paul applied different functions to the indiviual OFLs. You need to drop the file into the Input channel section than move it to the specified channel in Output Channels. Tyler talked about it here. SBench 6 has a Base version and a Demo Pro version, we only get 30 startups in the Pro version and we currently are at 10 left. I wanted to see what i could do with only the Base verison, which is onyl running existing projects or simply functions. I will probably need use some of the Pro attempts. I should be able to write in my own complex function to scan the frequency. Here is a video series to an introduction to SBench 6. Video 14 dicusses Complex function generators.

channel 1 from spectrum card → OFL1 RF_Amp Input

OFL1 RF_Amp Output → OFL1 AOM

ADWin0.DIO1.DIO29 → OFL1 RF_Amp Switch

ADWin0.DIO2.DIO15 → OFL1 Red Pitaya (dio6-pin & Gnd-pin)

OFL1 RF_Amp VVA path(OFL1 PD) ← OFL1 Red Pitaya (Labeled 172.16.74.40) (in network Sheet its label 172.16.74.149)

OFL1 Red Pitaya (OFL1 SP) ← AdWin0 (3-5 OFL1 SP)

OFL1 Red Pitaya (OFL1 PD) ← OFL2 Photo Diode (OFL2 PID)

direction matters….

I figured out how to output a signal through SBench. I launched the SB6 file call 698.sb6prj then i disabled all channels except for CH1(this is for the OFL1/698 AOM). I set the output to 100mV to be safe, Press Singleshot → Force Trigger. If you get a connection error, reopen the file.

I believe Paul used python to set the the AOM in the sequence,or now i need to just align the aom to the atoms. The output voltage of the Spectrum card are specified values. Here there are with there corresponding power(dBm)

the OFL1 AOM is 80MHz at 0.3W or 24.8dBm

Paul's OFL1 RF powers

before i had checked the power out of the spectrum card, I check th RF_amp output(w/ 30dB attenuator).My reasoning behind not checking the spectrum card power first was ,if i start at the lowest setting in SBench and and increase the Voltage in the smallest steps, when the RF_amp output is 0.3W(AOM Recom), i should be okay, since no one has touched that RF_Amps. I started at 100mVpp and saw nothing, at 200mV I saw a little peak. THan i increase it to 500mV, 1000mV,1500mV, and 2000mV got me to almost exactly 0.3W. After word Khang said I need to check the Input power. He wondered if i had broken the RF_Amp becuase the max power into should be 15dBm. I am concerned too, but i did start at the lowest possible Voltage and continuously saw little to no power out. I check the RF_amp output power at 500mVpp and it was -17dBm, which is a 20dBM drop from the input. I should contunie to tcheck for Paul's spec on the input power to the amp.

NOTE: While using SBench, i will only put 500mVpp for the 3.3dBM of power because Khang said it should be between 0-5dBm.

I learned how to generate a serivce report for the 689 topica laser. Download TOPAS 2.4 .exe software from server. ONce done , connect to the either 689 or 698 laser( this will not disrupt the laser). Go to Meni → Device Config → Misc. → Generate Service Report. It can only save as a .BIN file.

You can get a service report by the Controller. From the Home Screen → tap Information → Generate Report (I never actually did this, but this instructions was sent by Toptica)

Took pictures of the 689 beam

Tried working on the Sbench before running the MOT with grady. Still no power to the AOM but ill get.

Grady noted that the 4611 laser is 10mV higher today from yesterday. OUr RMOT OD is 1.87 vs yesterdays 1.47. NOt entirely sure why that is, but we will take it. the ODT ball was lower but grady noted that i moved the ODT1 lens in the wrong direction. HE also adjusted the ODT2 lens Depth to move its focus ( which helped it. The new ODT OD at 1ms_tof/1s_ID is 1.3

Grady moved the focuses of the ODT beams to get the ball closer to the center of the mot, becuase we know its aligned with the downward florescence beam. I removed the single mirror in the way of the EMCCD and attempted to image rMOT on that camera. I dont recall ever seeing the Rmot with that camera, and i thought the flipper0 might be in the way of the imaging. so i adjusted the imaging delay to from -440ms to 1s and i didnt find the mot. i had the gain at either 250 or 350, because we only needed 50 to see the ODT before. I saw tried moving the By shim coil from 1355 to 1485 mA and still no luck.

Grady got he rmot florescence on the CMOS. I moved it to over lapped the downward florescence beam via shim coils.

Afterwards we, moved the objective down to makes sure out florescene beam does not change

I moved the ODT beam to the new Rmot position. While ODT1 went smoothly, I cant not say the same for ODT2. Bx shim coil had the largest change to the odt beams. I was imaging at 1ms TOF, 1ms image delay. During my change from 182mA to 40 mA on Bx, ODT2 slowly vanish and was gone at 60mA. Grady helped me get it back. He started to adjust the Vertical know on the last odt2 mirror, and i was able to get a small ODT Ball at 1m_TOF/1ms_ID at 0.75 OD. I moved the move the last ODT1 lens closer to the window and adjust the last mirror of ODT2, and got a OD of 0.95. Then i went back to the normal Rmot and only adjusted the vertical Red beam. This got me a ODT Ball with OD 1, as seen below. I also took an image of ODT2 to see how it changes for tomorrow.

Shadowing grady:

we lose too much 461 power when splitting to both the upward and downward paths, So we change the but-coupling from one to the another.

he takes a 20 kinetic series images and export the .dat file to Stronitum/data/tweezer_alignment_'Date' and used the jupyter code “tweezer_alignmetn_497” to subtract the images of the upward and downward paths

When there is too much outside light, the gradys code is unable to find the “region of interest”, This part of the code is helpful to determined quantitatively how far the 2 beams are from each other. The will add the lines that will be go through each beam and be parallel.

grady is checking the input 497 and 461 imaging paths to see if its the input are are angle to the objective. He have a mirror reflecting back the both beams. We aligned the downward bImaging path and backreflected it off the mirror and made is as centered to the objective lens as possible.

We switch the mirror for a frost pin hole , to better align the downward Bimaging. the image on the camera looks a lot better and better parallel with the upwards beam.

Monday: grady got the beam to upward 461 to hit the cold atom beam.He also got a 2nd transverse imaging path, He saw the 461 was unlocking a lot, so he increase the power

Tuesday: I spent the day watching grady. We got the had los the upward 461 beam hitting the cold atom beam, but after some troubleshooting. A card that blocked the Zeeman light was removed, once put back we saw the bea again. We use ray tracing program to understand better how we could align the objective with the window. We determined that aligning a downward beam and upward beam through the cold atom beam could work. He adjusted the len configuration to 1x 400mm lens and 1x 400mm achromtic lens. Wehn adjusting the aligning of the new imaging path, we would start with the steady state Bmot floresence, then move to cold atom beam.

Wednesday: Grady got imaging of the 2 overlapped 461 beams, and proceeded to align them. After discussion with Julio, he took of the objective to see where on the window it hits. After centering the beam on all the mirrors the beam doesn't not align up with the window.

Thursdays: grady suggests we align the 497 beam, without the telescope lens, to the window than add the objective lens and back reflect at each set.

Sunday: Grady removed the mot lens and the telescope lens to the 497 and imaging path. Started the process to back reflecting each path off the window and centering at every mirror. Made sure the z-adjust to the objective would affect alignment, and use tip/tilt to align. he added the telescope lenses back to the 497 and imaging path and aligned it to the reference beam.

NOTE: the Tip/tilt, 497 and imaging are copropagating and shouldnt be touch. we will move the atoms to the best spot.

Currently grady is back propagating the new imaging path. By using NearNear-FarFar through the centers of the input and output windows.

HE replaced the Lens1: 300mm 2inch A-coated lens with a 400mm Achromat 2inch A-coated lens.

And Lens2: 400mm 2inch, no change

We are imaging on the CMOS Camera, with the new transverse bImage path. turned on BMOT florescence and got this:

Then changed the light on the atoms and lens position to get the atom signal brighter

Then imaged With the CMOS, the upward vertical bImage, and turn on cold the Cold Beam Florescence

so something was bumped and we dont know what tho

We went back to steady BMOT, but with no upward or downward beams. and looked at the fluoresces

There are 2 transverse bImaging paths, the normal one and the new one. You can go between them by removing the magnetic mirror right before the CMOS camera

NOTE: when imaging the steady state bmot, GAIN should be 0. When imaging the cold beam, start gain at 0 than increase it.

during the steady state, grady tried to walk a better image onto the camera

NO Luck

Tried realigning the the New bimage through the center of each lens. and went back to cold beam

Found it but it was faint. Make sure the gain is 2.6k+ and resale the image

found the problem, the cards stopping the majority of Zeeman slower light was removed, so we saturated the camera and couldnt see the faint cold beam image.

We put black tape onto the window where it was coming come. (Black tape on window)

Now grady put a tube on top of the objective with 2 back to back mirrors.

We seen the a 2nd beam go through the objective twice and hit the cold beam

We brainstormed how to make the objective paralell to the window. and after a lot of thinking came up with the following:(biggest challenge is divided the bImage power bwtn both paths)

align the upward beam to be back reflected off the window and roughly be going through the objective.

send a beam down through the objective and align the focus with the input beam going upward

then tip/tilt the obejctive until the back reflection of the downward beam is aligned with itself

Set of the 994 pick off to the wavemeter, but the output power is too low. I need to couple it better or change the collimator.

The OFL1 is back to adwin29, and the ADOXY is on adwin14(replaced OFL2). I find to learn how to use the S-Bench on the tweezer imaging computer.

Grady is using the cold beam to align the objective lens to the window. He set up a imaging path through the window next to the 481. and he images florescence at the CMOS camera.

I got the 698 through the fiber at 44% eff, with output of 56mW. It goes through the aom with an insertion loss of 6.5%.

I learn that the swofl1 was taken for the sw497AODXY. So i have to find a differnt Switch to used

I set up the Objective actuators, but have not turned them on. Im using ethernet port B0210-R1-B41.

The controller is currently connected to the third actuators that is currently not on the tweezer board.

I got the better Spectroscopy data,

I read in Tyler's notebook, that he only needs 3mW at the atoms for 698. With rough measurements, we should need 5mW out of the fiber. I currently couple 9.8mW of power out, but the beam doesn't look good.

It look like this with out a cylindrical lens

and this with

after turning off the lights and monitors, and covering the 461 ND filter. I still have a min noise on the camera of 520 counts.

holzworth: 330.223 kHz → 330.262 kHz I was able to change the holzworth but finding the brightest ODT on the floursence camera.

set up the new SMA cables for the 497 aod.

testing the old sma cables: power out of the x-port computer for 1 peak:

Initial: -2.18dBm

698/497 aod Y: -3.24dBm, no change in power when bending cable

No Label: -2.9dBm, no change in power when bending cable

497/497 aod Y: -3.24dBm, no change in power when bending cable

497/497 aod X: -38.50dBm, can jump -3dBm when bending the cable (broken)

I notice the X-Axis for the 497 AOD is centered at 84.07MHz, i need to ask grady about this on monday. And i also notice some weird small peaks after the RF amp on the same x-axis. Ill go through the whole Sma set up and figure out if its coming from the the cables, RF amp, and the weird 84.07MHz center.

Here is the manual for the spectrum analyzer( integrating total power)

only change the target wavelegth to at lowest 993.8 nm on the 497 soltsis

http://etheses.dur.ac.uk/12905/1/thesis.pdf?DDD25+

the tweezers are centered at (488,518) and the best center for DFG is (597,551)

Grady and i trouble shot the alignment of the DFG with the tweezers and the imaging path.

I put the DFG over the tweezers, i saw a 1×1, 2×2,3×3, i was able to see a 4×4 after 10 averages: the set up was 496.994870nm, output pd 0.874V, ecd output 1.77V, set point =0.12, 0=TOF, 5sec Imagedelay, 150ms holdtime

When adjusting the Hortizontial knobs of ODT 1 i noticed the top left corner and bottom right corner would change in population. and for ODT2, the top rom and bottom row would change in population. Has you can see from the figures, the bottom left tweezer was always less populated from the rest of the 4×4 array. I tried aligning the horizontial dfg to fix this but couldnt. could try vertical next.

i attempted 5×5 and didnt get the whole thing

spent the whole day setuting up 497 and 994 light to the wavemeter. i couldnt find the fiber thats specific to 944, which is a 980 fiber

when i when runninng spectrocopy on 497 (changing DDS to/from 162MHz), to save the DDS file and put it into the background. Once you have made an edit type in “ :w ” hit enter, and hit “ Ctrl-z ” to get back run ' fg '

i submited my nsf today, which was nervous racking

Trouble shot the elongation of the tweezer array and the Extinction ratio of 497. Khang said when we turn on and off the RF to the AOM, it changes the polarization of the light coming out. so grady implemented a way to has have the AOM on in hopes of fixing this issue.

THe tip and tilt knobs for the objective were adjusted to minimize the elongation of the array, but one of them ran out of travel. A way to get this travel back is to change the angles of the mirror above the objective but we dont wont to do that. So instead, grady went back and forth between the tip and tilt to get it the best and realign the DFG to be at the center of the tweezers.

The elongation got better, and Grady thought of adjusting the powers to the RF amp, to further minimize it, and its it helping.

I will have to redo this when i realign the 698 tweezers through the objective later in the year. becasue i would have to attach a beam on to the objective for alignment.

Make sure you understand how to got from NxN, NX1, 1XN, and YN, NX arrays

Turns out the OD drop was by magic, becasue grady got the OD to 1.7 by just lowering the powers, which makes perfect sense….NOT

the OD is at 2.2 and at 70% lower 689 power, which i wont be complaining about.

here is when Paul see a clock transistion: here

here is when tyler locked the 698

Here is the 698 clock laser page

The Physical manual to the to lock the 698/689 laser and the entire setup is in the vestibule draw, and the PDF of the DLC pro is in :/strontium/equipment/698laser/

I spent the day troubleshooting why the mot was so low (1.0 OD) Since neither I or Grady worked on the tweezers for 2 day, I believe it was drift of some kind or a stetting wasnt changed back. When i saw the OD of 1.0 i thought it could be a shutter was left open or the laser was detuned too much, but all shutters were closed, and i change the detuning of the bimage. I also changed the Holzworth and detuned the bmot with no luck to both. I noticed the R87mot and Rstir were not reaching set point and the monitor noise for locking looked too noisy. I proceeded to re couple the Rmot fibers to reach set point but no luck. Throughout the day the OD kept dropping, but i believe this extra drop is due to some misalignment.

Too make sure, i checked all the powers of every beam and checked the Florescence BMOT and i believed i could rule them out

I made a table that shows all the Efficiencies of every major fiber on the Clock tweezer experiment. Its in experimental_log

I will check alignment on monday since its currently 9pm, but i believe the 689 is the reason for the drop in OD

I mapped out a path for tth3 698 laser to go transverse to the atoms, and i made a list of opto-compontents that we will need and already have.

I transitioned my time to seeing if we need anything for a potential Double pass/ laser locking path. Grady arrived, and spoke to him about my finding. He lead me to Pauls thesis, which had a path for 698 laid out. The 698 shared the same path as the OFL and there was already a path for the vertical beam as well. Once i check all the opto-components on that path, i realized we should have everything we needed.

I need to double check Tylers and Pauls notebooks' to make sure im not missing anything. and learn how to lock the laser 698nm

Papers that will be helping in clock spec for Sr:

Thesis from Schrek group

Sr Clock Japan

note for future self, when dealing with very cold atoms, ramping the light power might be better because the light you are using to trap with might be heating up the atoms too quickly.

Has Grady worked today, he noticed an inbalance of trapped atoms in 2×2 array. some potiental reasons are: The powers of the RF peaks might not be equal make the powers of the indiivdual tweezer different, the DFG might be too small to equal load the same number of atoms into the tweezers, and the 497 or objective might be tilted, and do we need more power(because we are within the power range from theory).

I have been learning about magic wavelngths and how to measure a atomic clock. I found this sort of check list for an atomic clock from Boyd Thesis, pg57:

For the lattice confined neutral atoms to be a practical clock scheme, we would like it to satisfy a few key criteria at a trap depth UT =U0:

1. λm is a practical wavelength, and the required frequency stability on the laser is reasonable.

2. One can perform spectroscopy in the Lamb-Dicke limit (νT rap > νRecoil) and in the resolved sideband limit (νT rap > γClock).

3. One can efficiently load atoms into the lattice (U0 > kB T )

4. Negligible scattering rate for lattice photons at λm and U0.

5. Negligible polarization dependence of the polarizability at λm and U0

6. Negligible quadratic Stark shift at λm and U0.

The only way to sort out the practical trap design issues is to directly calculate the ac Stark shift and magic wavelength for our lattice configuration.

So let’s get to it!

i spent the day reading papers on optical atomic clock and measure magic wavelengthes.

While the info i learned has not yet solidified, the papers i reviewed were:

Dan Steck Quantum optics

magic wavelength by barreiro group

Boyd thesis

I also set up a code to play with Trap depths and light scattering, Its located in /strontium/code/Tweezer_Calc/Trap_Depth/

When you want to run Jupyter Notebook on a differnet moniter, go Here in the Wiki

I reset the DDS blue script and checked all the RF powers of the all 3 tweezer array codes(Gradys new code, Old code, phase update code).

I played with the Repumper and noticed it was is multimode. Look at Pauls thesis to reconginzed the difference.

I was able to increase the OD to 2.3, at 481.321900nm at single mode.

While checking the Sequences of the old Array code, i noticed a 0.2MHz shift on the center frequency for Seq 1,4,5 And no shift at Seq 2,3 Seq 6 has no output

I also measured the powers of different arrays: Grady's code(GC), Old Code(OC), and Phase update Code(PU)

GC single tweezer: 30mW GC 2×2: 17mw PU 2×2: 3.5mW PU 3×3: 2.2mW Pu 5×5: 1.2mW

changing the DDSs give you the ability to detune the Red and Blue light of specific locations

In a terminal:

type: ddsblue

pwd is found in /common/networking/vlan.xls

to edit the code: type: vim run_blue_DDSs.py

Scroll to the bottom, to line “####DDS3” change: “ DA3 ” from 177 to 162 and reverse 162 is the center frequency of of the bimaging aom, and changing it to and from 177 is actually a 30 MHz change, because of the double pass the light goes through.

ex) 162 to 168 is actually a 12MHz and not a 6Mhz change

Once your done editing, adn ready to re-run type: python run_blue_DDSs.py 87

Commands that will help you navigate and leave a specific situation:

to quit, type : Shift z z

to bring something to the forward ground, type in the command line: fg

to detach: screen -rd blue

so we got image of the the tweezers and an image of 2×2 array. We are unsure if they are single atom images. THis imaging was done with grady's 500mm lens in place. (mag =500/16=~30x

We tried switching to the tube lens Mag=180/16= 11x, but we notice the ODT in bigger on the camera and we dont even see the Tweezers anymore. There could be 3 potential issues;

1. objective is not in focus at the tweezers and ODT.

2. The tweezers and the ODT are aligned with each other.

3. The imaging system in not in focus at the camera.

4. Something else we have yet to think of.

The last thing we want to change is the objective , because we get an image so we choose to turn off the 497 and optimize the ODT image.

We did this by changing the tube lens to get the beam in focus(high intensity), then we walked the imaging mirrors to get that image into the center of the camera. With the image position on the camera being changed with the closest mirror and the intensity being change with the farthest mirror.

The Aosense angle valve work prefectly, and we were able to heat straight to 480C with out the Aosense ionpump going to 5e-8torr

After Getting the 461 light back to set point , I tried to get the MOT back. After some re alignment i realized the Holzworth wasnt change from the weekend. (NEVER FORGET THE HOLZWORTH). That increased it, but not to ~1.48 during 3s load time. I switched to Florescence BMOT and optimized it. the key was the ZM and 2D MOT, they had the most effect. After i went back to OD imaging and only adjust the 689. I got it to 1.35 OD, but lost it slightly. Grady helped me get it back to ~1.37 OD. I learned, it very hard to change the 689 transverse retro mirror that is a flipper mirror, with a alan screwdriver. You need to use right angle alan key (1/16“) get a more precise turn. ( I left some already there).

YOU WILL NEED TO REALIGN THE ZM AND 2DMOT, becuase you applied heat to the whole aosense

But our MOT suck for some reason, while i was realigning the beams, Khang noticed the Output PD on the 461 SolsTis at 0.740V. It should be at 0.770V. BUT whats more important is the ECD output, it should be around 1.085V This would cause our 461 light to not reach their set points. The 461 laser is dying and there our 3 ways to and increase if your power isnt high enough. (target wavelength is 922.8nm)

So, as we both know, the alignment drifts and power drifts. The 3 steps to reaching setpoint of 461 is in order of what to troubleshoot first.

1. Make sure the power going to the microscopy strontium gas is at least 0.59mW or as high as possible (currently 0.723mW). Before you change any optic, check the power at every location. YOU CAN NOT MAKE A MISTAKE HERE. Optics to to try is the mirror just before the collimator and the Single pass AOM.

2. Next is to change the Output PD. You first connect the BNC cable labeled “AA Monitor” that is coming out of the sprout. The iceblock the assembly that connect from the big box above the laser table(the one that you turn the key to turn on the 461) and the laser going into the TiSaph. You take off the you close the shutter from the sprout and carefully remove the metal cover. You then walk the 2 mirrors to increase the Output voltage to as high as possible (currently at 754mV).

3. The last thing you can change is the Power of the Sprout itself. This is the last possible option. You dont consider this unless all previous steps are optomized. The sprout is currently set to 12W out its possible 15W. You should only increase it by 0.5W. The reason you change this last, is because running at a higher power will shorten the life span.

we checked the Aosense angle vavle was a good, by first closing the Valve to the main chamber then flooding the pump and larg bellows. We let it sit like this overnight. I went home early.

I had classes today.

After adjust the align of the rmot and bmot, i got a OD~2.0

this is what the BMot looks like at Rmot OD~2.0

The day before i got a decent enough MOT(OD 1.6) to start finding the ODT beams. I ended that day was 1 ODT beam. Grady pointed out the the intensity of the 2nd ODT beam was full the entire run time, but it should decrease(maybe the photodiode).

Today:

While setting up the photo diode BNC cable, i noticed the photodiode was turned off. and i turned it on and saw the odt beam.

I was able to align the the odt beams vertically , but they were already close. I slight moved their horizontal direction, but i was hesitant to move them far because i couldnt see how much i was changing them. THe rest of the day i thought of ways of using the vertical imaging path to image the florescence of the ODT beams on the cmos camera. I played with the sequence, but got no where. Grady explained how he got the ODT beams to cross. He solely use the horizontal imaging path to see a ball of atoms start to form in the center of one of the ODT Beams. I change the Image scale to match his, from a previous date, and i saw the Ball more clearly. Before I started the whole vertical imaging of the beams, I had already mostly aligned the Beams, but i wasnt aware of it due to my scaling.

I align the beams a little further and got the image below.

Holzworth: 330,085kHz → 330,113 kHz

Coupled the 3 way Bmot to a slight higher power. (made sure not to mess with the wave plate!!!)

I coupled the downward Bmot to a higher power as well, but i messed with the wave plate just before the collimator.

I had to do an Extinction Ratio. I got 7.8mW going into the fiber, 5.35mW(69%) coming out, and 2.75mW(51% =2.75/5.35) coming out of the telescope.

Focus bwtn both Axes:

X-Rays w/ focus 6.4mm closer to x-axis

Y-Rays w/ focus 6.4mm closer to x-axis

X-Rays w/ focus at x-axis

Y-Rays w/ focus at x-axis

X-Rays w/ focus 6.4mm farther from x-axis

Y-Rays w/ focus 6.4mm farther from x-axis

1. Close the OD Imaging screen of the on the imaging monitor

2. Trigger the camera 3 times from the Virtual machine

3. Connect camera to Andor Solis App.

To got back:

1: disconnect camera from Andor solis app

2. Run matlab Imaging script

3. Trigger camera 3 times

I aligned the new 497 set up, It should be easy to replace the older waveplate with the new one once it arrives. and further algin it with the ODT.

The purpose of this documentation is to add all additional procedures to do a successful oven exchange for the Clock-tweezers experiment.

Here is previous documentation for wrapping, and bakeout procedure

Here is info on the Oven itself

Starting at the setup, We made a connection from the turbo in 3 different ways: 1, RGA/experiment, 2. New oven, 3. the sieve. This is usually accomplished by a single cross flange, but we didn't have one, so we used two t-flanges. Each end had an immediate valve to control where we are pumping down.

For the Sieve section, we had to use 2 additional valves because we didn't have a single valve rated for high vacuum to atmosphere. Having multiple valves helped elevate that immediate pressure difference across a single valve.

The small Bellows that connect the RGA to the Aosense valve should have an elbow flange because this will allow the small bellows to be less in the way of the glove box. That elbow was not installed this time because we didn't have the piece.

1. Tape down thermocouples with resistant tape while prioritizing sensitive locations such as the turbo, RGA, Windows, and valves.

2. Wrap a single layer of the Vacuum foil, and make sure to have it firmly pressed against all surfaces.

3. Wrap Heaters around all surfaces while maximizing the surface areas covered. Windows should not have heaters over them; they should have a protective layer of metal mesh.

4. Wrap all surfaces with additional layers of vacuum foil to keep the heat in. This wrap doesn't heat to be as tight as the first initial layer.

Helpful notes:

To detach a screen, press control-a  then d.

To leave a ssh, type: logout

If you want to place the bakeout screen back onto the nanotrap monitor and you get the panic script screen, you need to restart the panic script and then proceed to restarting the bakeout script.

Open: XQuartz app(Mac only)

While SSH in nanotraps:

To get the bakeout on your screen

Type: ps -ef | grep bakeout

Type: kill -9 [PID#]

Type: cd src/bakeout/src

Type: conda activate bakeout

Type: python bakeout.py 

Type: python bakeout.py

To get bakeout on the Nanotrap screen

Close Bakeout screen

Type: screen -rd bakeout

Type: python bakeout.py

It should say "starting scan"

Check Stable Temp

Type: ssh tk

Type screen -rd

Type: python BBB_tkdb.py

Emergency Kill:

Type: ssh kill

Type: screen -rd

Press: Enter

To Raise Max Temp for Alerts:

Type: screen -rd panic

To exit, press: ctrl-c

Type: vim variables.py

Edit as needed

Type: python temperature_panic_check.py

There are 2 main gas lines that you need to be a aware of Special Gas and Nitrogen Lines. These lines wont necassary have the specified gases in them but rather they are just labels so you can connect them to the connect lines. For the the Argon Set up, both lines will have Argon in them. But the this oven exchange, Nitrogen line went to the sieve, and the Special gas line went to the Bag/Glove box.

In the room, where the chillers are located, are the tanks and valves to these lines. The lines are labeled. The valves on the tanks will be closed, but the regulator valve will be open and set. Open the tank valve to start the flow of Argon gas to the lines you will be using.

Above the Nanotraps table and near the 497 laser side of the Clock-tweezer side of the main table is the 2 Lines that we will be using. One labeled Nitrogen and other Special Gas. We know that both of these will have Argon in them.

Both Lines will have a regulator valve and a pressure sensor to them.

The Argon line the goes tot the sieve will have the Cut-Glove as its regulator and its pressure gauge be connected to a DMM.

Cut-Glove Regulator:

Pressure Gauge:

How to monitor:

The Argon line that goes to the GloveBox/Bag will have a similar set up but with a different gauge and regulator:

Gas Regulator:

Pressure Gauge:

How to monitor:

ONce you have proper flow into the Main lines, you will need to monitor both flows.

Start the flow into the glovebox, and make sure you have a postive pressure of ~4-5Pa If you dont reach that then you have too large of a leak in you glovebox, and you need to better seal it.

THen you will need to bring Argon into the chamber , Turn on the Gas lines that leads into the Cut-Glove Regulator, but DO NOT open the valve into the sieve. Make sure there is a small positive pressure of argon gas coming out of the cut-glove. The glove's inflation is your indicator of your postive pressure. you will begin to open the valves starting from the back of the sieve to the Aosense in these steps:

1. Make sure all valves from the sieve to the Aosense are closed

2. Make sure there is always postive Pressure of Argon flowing out of the Cut-Glove at ALL TIMES.

3. Slowly open the the valves, starting at the sieve, exposing each smaller chamber to only argon, make sure to look are the Cut_glove doesnt deflate at any time. ALso, look at the DMM to read the voltage does up the 7.5V, the is 1atm.

4. Do the same step for each individual valve leading up to the aosense's Angle valve. Becuase the Large bellow will take too long to fill with argon, fill the large bellows before you fill the New transfer CHamber.

5. Once you have the glove box and all valves(Not including the Aosense ANgle valve) at the proper pressures of argon. Slowly open the angle valve. make sure not to over pressure the AOSense. Over pressuring the chamber could cause the windows to pop off. so be ready to decrease the argon flow. Be aware that opening the Oven will close the pressure to decrease, and sealing the new oven will cause the pressure to rise

6. Now you will be ready to exchange the oven. make sure the DMM reads ~7.5V and the laptop says ~4Pa, while the oven is being exchanged. Once the oven is properly exchanged and sealed, close off the argon flow at the sieve Valve and start the Roughing Pump. Then go to the Turbo and do the proper bakeout proecdure.

I can use thermo:t20,b13,b3,b4,b9

b9 is absorption window 1

b3 is 2D not window

b4is absorption window 2

b13 is ion pump

t20 is bottom window

I know i havent put in an entry for 6 days, that is because i have been working on the glove box and cutting, changing the design a bit, gluing, etc.

The purpose of this entry is to tell myself to tap 1/4-20 holes into the floor of the glove box that sits in the 813 area. They should be about 3 inchs apart.

also drill a through whole into the same glove box to make sure you can put a 1/4-20 into the table. (i have 2 holes that are mistakes. no more mistakes!)

(Sept 4th to 15th)

Today was my first day back from my honeymoon/ being sick.

While I was gone, Grady got Tweezers for a single day and the next lost the MOT. We are unsure why this happened but the most likely culpert is the oven.

MY task for the day was to measure the polarization of the beams and the wave plate markings. Note: Due to space, i was not able to measure each of the polarization right before the atoms and some of the Wave Plates:

The 2D_MOT #1 was non- circular and non-linear at the atoms. I was not able to measure the polarization of the 2D_MOT#2

This is the polarization past the second window of the 3D_MOT#2

Regarding the RMOT: both transverse beams are Left Circular Polarized at the atoms. The Verticals beam were not measured

from the 10th to the 12th, grady and i have been trying to find the 497 tweezers with either the OD from CMOS or the florescence from EMCCD. The effect of the tweezers is extremely small on the MOT, so we resorted to making the waist larger. WE do this by decreasing the size of the beam going into the page ( using irises). THis will also lower the power at the atoms, which we dont want.

Today(12th) Grady had the idea to put 2 irises in the center of the path of the 497 tweezer so we know where the tweezer goes on the top board. Then we put the 497 fiber into a new separate collimator (cfc5) and aligned it through the iris. This setup gave us a direct 497beam into the objective that would be 1.5cm diameter and have up to ~600mW of power.

We were about the start imaging, but Grady wanted to make sure we were going through the atoms, so we checked the alignment of the new 497(small) with the upward BMOT, and we were not aligned at all. We decide to alignment the 497 with the Upward BMOT, and noticed it was far off the original 497 tweezer path. We aligned this by doing near near far far. Near would be the kinematic mount the collimator was on and the bottom window of the chamber, Far is the mirror we added to the path and the input BMOT on the BIG Table under the chamber. Once we got this close, Grady started scanning for the tweezer.

We were putting 40mW of power at the atoms with a Input beam of ~1.5cm diameter. When Grady started scanning, he assumed the focus was aligned vertically with the MOT becuase we were the focus using the florescence using the EMCCD. and he assumed we were aligned side to side. He scanned only the Bx-shim(in and out of the page) and found the image below.

The tip of the MOT is cause by the Stark Shift, Which i need to review more of. (paper on magic wavelength, martin boyd thesis, stellmer)

We moved the MOT to see if we could characterized the 497Beam, by either adjusting the SHIMS or by doing Time Of Flight(TOF). WE notice we see more trapping when we drop the MOT above and through the focus(image below), in comparison to dropping the MOT at and through the focus.

1.5mm waist and 60mW input:

We also adjust the collimator and mirror to see if the focus changes position vertically and how the beam changes directions. When Adjusting the collimator, the focus 'maybe” changes 70ums vertically, and the mirrors dont dramatically change the direction of the beam.

WE switched the CFC-5(1.5mm waist) to a collimator with 3mm waist to get a bigger beam, and see how that effects the MOT and how repeatable it is. The beam direction is repeatable but the trap becomes smaller or not there at all, and atoms are less likely to to stay in it.

3mm waist and 60mW input

This is how we some 497

Holzworth: 329,728kHz → 329,732

Holzworth: 329,724kHz → 329,728

I ran at 55mW of 497, and didnt see much. Grady says to take multiple sets of data and subtract them.

Because we plan to put a lot of 497 power to the atoms, we switched the fiber to a high power PCF Fiber. I was tasked to get the coupling to 40%, which i got but it took a while. I realized i need to profile the beam before and had to redo the coupling. At this point grady got here and we switch the “ proper” pcf colimator, but it prove not to work as well. We went back to the CFC colimators and got 50% eff with a CFC-11.

We found the Extinction ratio would be messed up once we change the Pump power from 10W to 15W. And that it would need another extinction Ratio after about an hour.

I aligned the Optical Isolator and profiled the beam. It was determined that the beam was being distorted by the Isolator. With grady's help, we found that one of the PBS in the isolator was causing this. He decided to completely scrape that piece and use the waveplate/pbs (beam dump) as the first isolator PBS. This fixed the shape of the beam.

Optical Isolators are PBS → Faraday Rotator → PBS. The first PBS changing the polarization of the light to align it to travel through the Faraday Rotator. The Faraday Rotator is an strong magnet. The second PBS changes the light to minimize travel back through the Faraday rotator.

We plan to use about 400mW of power on the 497 Tweezer. so in order not to burn anything, I put 497 AOM in the sequence to be off normally and turn on when the RMOT is loading and turn off immediately after. The AOM is called 'Sw497'

The 497 laser has an Optical Isolator. While i dont know what it does yet, it is extremely dependent on Wavelength.

MAX Power out of the laser(15W) 940mW:

Aligning:

Power out of the laser(10W) 360mW:

Eff of the Optical Isolator: ~86% (313mW)

Power range of the waveplate: 0.3mW to 294mW

insertion loss of the AOM: 6%

Eff to the 1,1 mode:

Eff out of the Fiber:

Gained More Experience with the Sequence:

The Main run sequence is the file 'main.pys' located in folder :Home/sequences/

The Channels that go onto the main run sequence is the file 'Initialization.pys' located in folder :Home/sequences/

TO Add a Channel, you add it in file 'channels.py' located in folder :Home/sequences/config/

If you want to add a Channel. you Must restart the Gui(QControl)

1. In Terminal1: go to folder '/home/lab/qcontrol/' , and run pyro-ns -n 127.0.0.1

2. In Terminal2: go to folder '/home/lab/qcontrol/' , and run pyro-es -n 127.0.0.1

3. In Terminal3: go to folder '/home/lab/sequences/ , run ipython , then %run -i TimingServer.py , then tc.startOutputProcessSingle() , then tc.startOutputProcessSingle() , then %run -i ../qcontrol/QtClient.py

I float main clock-tweezer table. The weight is ~3200lbs for 41psi. I adjusted it to be level. all info is found in the OptoTable part of the wiki

Grady and i went through different steps to find the hole in the MOT due to the tweezer. I spent the day adjusting the AOD code to get 20×20 grid of tweezers within the 50MHz bandwidth of the AOD.

RF_Amps_Page

In our RF-Amps we have 3 individual Amplifiers and 3 ways to attenuate those Amplifiers.

Now referring to the image below:

We use the Code to drive the initial max voltage input into the RF-amp. That max voltage is determined to be ~80% of the max voltage of RF-Amp1 (ERA-2).

The Switch(ADG901) has 2 settings; Manual and BNC. It is currently in Manual and in the Always On position. We will later switch it to BNC and control it through the DIO.

The RF-signal then goes through a Digital Step(DAT), which will lower the power, specified by which dip switches are ON. Then it goes through the MAAV(Potentiometer) and then RF-Amp2(ERA-5).

Finally it goes through RF-Amp3(RFC1), which has its know attenuation built-in.

1. When connecting to the RF-Amp, connect the OUTPUT, then POWER,and finally INPUT.(disconnect in the opposite order) This is extremely important and improperly connecting it could result in breaking the RF-Amp.

2. Find the max voltage allowed by RF-Amp1(ERA-2) and set your code to allow no more than 80% of that voltage.

3. Set Dip Switches -16dB and -8dB to ON

4. Set Potentiometer to is maximum output position.

5. Adjust the Dip Switches until you get (the power specified on the spec sheet)+ (0.3W or 0.4W).

Helpful Table of dBm to Watts to Volts

the code to run is ./rdma_fifo_kernel_DA

located in file tweezer@tweezer-desktop:~/scapp/cuda_rdma/rdma_fifo_kernel_DA_20230606/ 90 90 0

the input power to the RF Amps is -2.5dBm or 560mW

The Output power after the RF Amps according to spec sheet is 0.7W or 28.5dBm @488nm

y-axis potentiometer max 30.68dBm or 1.17W

x-axis potentiometer max 30.75dBm or 1.18W

i set up the tweezer camera. and learn more about the sequence, and how to add to it. I imaged the fluorescent of the atom with the vertical imaging. The only issue was i had to use 1 lens and the image was not good.

I set up an easier way to keep track of the powers Here

With the original Blue MOT telescope lens in place, the vertical image is a lot better (more of the light is reflected). But the main issue still stands: the light is still focusing near the atoms and diverging out of the chamber.

i tried to image the atoms without the second lens in the vertical imaging. I took a lot of time but I got it , but it wasn't consistent. It was at too low of power, so I quickly coupled the fiber from 23% to 33%. and it helped.

I also noticed the downward BMOT was not coupled efficiently, so I improved that from 22% to 32%

With the vertical imaging, without the 2nd lens, the atoms are not on the focus of the camera and the image is bad.

Grady and I set of the 497 RF amps and the Motorized MOT lens, and started to look for signs of the tweezers. SO far no luck, but we tried Delaying the image after the Rmot turned off, we moved all 3 Shim coils to move the MOT, we lowered the objective to be max 2mm from the window.

Grady and I optimized the MOT to OD 1.9.

We did this by looking at the repumped atoms and how they move. We determine Ramping the Z_Shim is the best thing to do b/c it bring up the repumped atoms that start to fall.

We replaced the 500m BMOT telescope lens with the original cage lens.

I learned about the timing of the shutters and how they work. If you need help remembering start Here

DFG Sizes

I measured the powers of the all the beams. There is a large power difference btwn the upward and downward BMOTs. There is also an overall low power at all the RMOTs.

Program(On the Dell): ELLO

When turning on the Micro-controller or if you bumped it and need to get it back to the optimized postion. follow these steps.

Referring to the image below.

In “Control”, Click 'Home'.

In “Parameters”, set 'Jog Step' to 1.52

In “Control”, Click 'Forward' #your position should change to ~1.52XX

If the Position is <1.5200. Redo the previous steps, because the program will give you an error when Clicking “Backwards”

I set up the vertical imaging path. I used 2x 300mm fl lenses, because the distance from the atoms to the camera is 1200mm= (300mm)+(300mm + 300mm)+(300mm)=(F1)+(F1+F2)+(F2). I notice it was light was focusing so made Grady aware. After looking at how the light is changing, we determined the light is not collimated , but rather focusing near the atoms, and the will cause the image of the atoms to be magnified. I need to find a way to image the atoms with a known Magnification with out the the light focusing.

Grady and I moved the ODT and Mot up within 4mm from the top of the chamber.

We did this but moving the Shim coils in the sequence, then only moving the vertical tilt of each ODT beam. We iterated between until it moved ~5-6.5mm.

The purpose of this is to move the atoms to be within 4mm of the top because the tweezer light will only focus about 4mm after the window.

This took the entire day, so I was unable to do the vertical imaging

Grady and I optimized the MOT by changing the alignment of the upwards Blue MOT, then the downward Blue mot, then upwards red mot, and final the downward redmot. We notice the Red mot had the most influence because the beams were smaller.

Also, since the downwards beams are so large before the MOT lens, they really dont change the alignment of the beam at the Atoms, but rather the intensity of light the atoms sees.

If the MOT is ever gone, put the the MOT into 2D Steady. The MOT could leave due to non-overlapping alignment, unbalance light intensity, and polarization.(order of importance/ most to least)

After optimizing the MOT to 2.0, i check the vertical imaging, and notice it was on the other side of the waveplate. I tried to get it back to the original side but it wasnt moving as much, so i will be picking off the light at the side its currently at and reflecting it across the table.

open the shutters on the laser table 2D mot, Zeeman Slower(ZS), and 3d mot

open the shutter for the upward 3dmot on the main table

in the Virtual Machine:

Set the BitterCoils from 65.122*uA to 41*A (UPDATE: 36.7A, 07/02/2025)

Turn On the MagFB_PIDTrigger

Took down the frame and finally closed all the doors on the table.

I notice the upper blue 3Dmot is misalign, so in curious if our 2od is from that, or we need to align it,

I started the blue vertical imaging path. The atoms are ~300mm away from the lower waveplate. so I added a 45-degree elliptical mirror to collect the light, and I attached it to the same 1-1/2 in post as the wave plate. there is an open hole through the main breadboard where the light will go through to the camera.

I will continue with the vert imaging path tomorrow.

we spent the day aligning the tweezer board with the chamber. we place the mirror, mot lens and objective, and got a OD of 2

FINAL MOVED THE TWEEZER BOARD!!

(7 units)

Note: C is Class related. TA is TAing. TA* is my sections for weeks 2,4,6,8.

I will be gone from May 30th- June 7th, because im getting MARRIED!:P

461MOT

diameter at the atoms are 6.1 +/- 0.5 mm

5mm away: diameter = 7.5 +/- 0.2 mm

10mm away: diameter = 8.8 +/- 0.5 mm

Right_hand Polarization: QWP 208 , HWP 157 Left_Hand Polarization: QWP 190 , HWP 195

689MOT

at atoms: diameter= 5.1 +/- 0.1 mm

5mm away: diameter = 5.5 +/- 0.1 mm

10mm away: diameter = 6.2 +/- 0.3mm

Right_hand Polarization: QWP 207, HWP 91 Left_Hand Polarization: QWP 206, HWP 45

497 tweezer

the first few knife-edges, were giving a waist of 400nm. even after adjust to the lens distance the lowest waist was 375nm. Grady remembered the input waist was not idea, so he check the ratio for measured waist:needed waist:: 1.05mm:1.4mm. this ratio is would connect to the lack of a 280nm waist because 1.05/1.4 ~ 280/375.

We adjust the determine the lens in the collimator was lens and that would explain the input waist size. but after increasing the input waist, the output waist decrease to 330nm ish.

497 tweezer: input power is 0.75mW,

insertion loss of the AOD is 4% (0.72mW)

Eff of the 1,1 mode 69%, after the insertion loss , 0.50mW ( which the spec sheets says >/= 65%)

the Eff of the beam slitter (0.23mW)

Beam Waist: 1.05mm

698 tweezer: input power is 3.27mW ,

insertion loss of the AOD is 6% (3.07mW)

Eff of the 1,1 mode 69%, after the insertion loss , 0.50mW ( which the spec sheets says >/= 65%)

the Eff of the beam slitter (0.23mW)

Beam Waist: 1.5mm

i went through the proccess of taking off the mirror/objecitve

461nm:

power after first telescope lens: 59uW

power after the objective: 15uW

25% PWR Eff

Right-hand Polarization is when HWP =167, QWP =242

Left-hand Polarization is when HWP =120, QWP =245

near the atoms the beam diameter is ~6.1mm, at 5mm away ~7mm, and at 10mm away ~7.8mm

689nm:

power after first telescope lens: 3.9mW

power after the objective: 3.1mW

80% PWR Eff

Right-hand Polarization is when HWP =109, QWP =216

Left-hand Polarization is when HWP =152, QWP =216

near the atoms the beam diameter is ~4.6mm, at 5mm away ~5.2mm, and at 10mm away ~6.1mm

MOT Size Comparsion;

689MOT(note for 689Mot, i havent added the HWP, soi will not add markings to the cage rods until during the final alignment.)

At the farthest telescope lens position: the diameter of the MOT is ~1.6mm +/- 0.1mm

At the closest telescope lens position: the diameter of the MOT is ~8.4mm +/- 0.3mm

461MOT(

At the farthest telescope lens position: the diameter of the MOT is ~1.85mm +/- 0.05mm

At the closest telescope lens position: the diameter of the MOT is ~8.4mm +/- 0.3mm

The final 3“mirror/ objective was was partially misalgined, I moved the micrometer of the Mirror away from its orginal postion of 16/30.

I put a beam profiler at the waist of the 497 tweezer, and then adjusted the MOT telescope lens to get a 6mm diameter at that location. The 2 mot images below are with out the 85mm MOT lens.

689 MOT:

461 MOT:

Holzworth: 329,563 kHz → 329,571 kHz

file:689/2023_05_17/take1

file:689/2023_05_17/take2

I move the objective lens ~1mm away from the camera:

file:689/2023_05_17/take3

file:689/2023_05_17/take4

I moved the objective lens ~2mm to the camera

file:689/2023_05_17/take5

file:689/2023_05_17/take6

These positions have been marked on the cage rod in green, so we know the range to place the lenses. But just to have a peace of mind, I moved only the 75mm lens 1mm away from the camera.

file:689/2023_05_17/take7

file:689/2023_05_17/take8

Now i only moved the 75mm lens ~2mm towards the camera

file:689/2023_05_17/take9

file:689/2023_05_17/take10

I drew the possible positions of the imaging telescope lenses. The blue lines are for the 461 imaging and the green lines at for the 689 imageing. The distance the 689 waist will travel is ~200um and it has a magn of ~90x.

file:461/2023_05_16/take1

file:461/2023_05_16/take2

file:461/2023_05_16/take3

For 461nm Takes1-3: the average waist is at (66 +/- 6)um, During these takes the lenses placements were not changed.

After moving the small objective ~0.5mm closer to the camera and still within the blue lines:

file:461/2023_05_16/take4

file:461/2023_05_16/take5 \

This slight offset gives raise to average waist location at 64.3 +/- 7 um, This hopefully shows that movement of the small objective does not effect the the waist position in a major way.

I changed to the 689nm,

689nm initial placement is about 40um away from 461nm

file:689/2023_05_16/take1

I moved the 75mm lens ~1mm away from the camera

file:689/2023_05_16/take2

I moved both the small objective and the 75mm lenses so the smallest imaged for 689 was around the 69um(16mark on the micrometer). These positions are still with the the red line marks on the cage rods

file:689/2023_05_16/take3

file:689/2023_05_16/take4

file:689/2023_05_16/take5

file:689/2023_05_16/take6

No lenses were moved from take 3-6. Because takes 4-6 were consistent in their waist location, i believe there was an unplanned movement of the lenses from take 3 to take 4. So i will take the average of takes 4-6 which is 34 +/- 7 um

I checked if i could repeat the 116x Magn, and i can i got ~110x on the 690. the cage will be marked red for this Magn.

file: 2023_05_15/take1

I went to move back to the 187x configuration for 461.and got 158x.

file:2023_05_15/take1

I retook the data for 461nm, after a small against meant to the angle of the small objective lens on the cage. (its still between the 2 blue lines that Grady drew). and i got 186x Magn without omitting any points

file:2023_05_15/take3

I went back to the 689nm configuration. and got a Clean ~120x Magn. The waist was measured at 77um (too Far)away from the 461 waist.

file:2023_05_15/take2

I moved 75mm lens ~1.2mm closer to the camera, where its in between the 2 red lines on the cage. and got 78x Magn and waist displacement of 51um

file:2023_05_15/take3

i moved the 75mm lens maybe 0.5mm away from the camera and still within the 2 red lines(i could not perceive the distance, it was by feel). And got a 96x Magn and waist displacement of 23um in the opposite direction .

file:2023_05_15/take4

I took a reference of 689 Magn. Its still at 85X.

file:2023_05_13/take1

I only moved the tube lens closer to the camera from the reference, and got a Magn of 70x

file:2023_05_13/take2

I only moved the tube lens farther from the camera from the reference, and got a Magn of 82x

file:2023_05_13/take3

I only moved the small objective lens ~1mm farther from the camera from the reference, Magn is 116x. Also i added in “Positions Moved” to show where from the original we are at

file:2023_05_13/take4

I only moved the small objective lens ~2mm farther from the camera from the reference, Magn is Trash. This is due to the diffraction rings on the images

file:2023_05_13/take5

I spent my time going through what Grady has done for the knife edge and waist generating codes( With Grady's Help).

I did a knife edge on each of the 461 and 690 to make sure they were in still overlapping, which they are.

There was an issue i came a cross; During the 690, The Fit curve for each slice did not reach the ceiling of the data. This was solved by Grady's input of adjusting the slit_width code to be a function of a tilted angle, because the slit currently is not on a tip/tilt mount.

When it comes to the waist/magnification code, I need to worry about the first 5 cells of code. They will give me the x and y waists at different positions. and then cell 6 will give me the magnification of those positions.

To explain this further;

The 690 and 461 lasers on pointed vertically on a 3-axis actuator mount. the only actuator that will be adjusted is the z-axis, the other 2 are aligned through the objective and dont need adjusting. When you adjust the z-actuator, you move the waist of the 461/690 in and out of focus of the objective. Grady marked the side of this actuator with a blue pen, because the actual reading of the actuator is on the back and it is hard to see. we take an image of the laser, at different z positions. the first 5 cells give a x and y waist at each z-position.

The data below is of my first attempt of taking a magnification measurement. This was using 461 at at the lens position for 690 85x mag. I took 14 images, starting where the blue mark was at 13 and moved down to 0. The only significance of this was to learn what to do.

I notice the code only allows for 11 data points, but i could be reading it wrong. also i need to start at the smallest waist and go up from there. So probably moving from 0 to 13

2023_05_12/take1

motor mount lens before the objective.

I scratched the lens so we need to determine which one it is. It is not labelled, plus it actually doesnt work.The MOTs still come out converging.

to determined the proper lens, i will use trial and error

30mm lens: converges

40~45mm lens: converges

50mm lens converges less

unlabeled lens (>50 mm) : converges less then 50mm

lens that “labeled” 100mm(might not be 100mm): comes out diverging

125mm lens: diverges more then the 100mm

75mm convex lens gives the least converging mot out of the objective, where is

80mm achromatic lens gives the least diverging mot out of the objective.

the 75mm and 80mm are slight too large for the mount so they couldnt be properly place in

497: power:767mW , Output:835mV, ECD Output: 1.34V

worked on aligning the mot with the telescopes and aligning the vertical imaging through the telescopes

The imaging is not getting better, and after discussing with the Grady. the plan to to make 461 the main reference path.

1. Get the 461 imaging to be the same magnification as the 689 imaging by placing in a larger focal length lengths.

2. Determined the position of the lens on the imaging cage of the 689 in reference to the 461.

3. align all tweezers/mots/blue Vert imaging path

4. align the waist of all tweezers with the 461 imaging focus.

5. Check the Mot positions through the objective

focus of 461 is currently at 6.1287mm, but this will change

Reference before the large focal length lens at 260X Magn at column 7 row 3

690 reference at column 5 row 4

i turned on the DFG again today. Some things to remember are to make sure to click Camera Shutter 6 times when going between horizontal and vertical imaging

I decided to check the clarity of the images with the test target at their individual focuses.

With 461nm at 6.5420mm, and ALL lens in place, the following its the clearest image i can get. the Image is of Column7 row6. which are 2.2um thick. We can still see the diffraction lines.

I notice these ring diffractions line mean im having only a sections of the line going through the small objective. I can move the imaging mirrors to the light to be at the center of these rings(where there is not rings). But at the center is all the light, then it we becomes too bright for the camera.

i tried to change the test target to the waist measured and adjust the small objective lens, but the image got completely taken over by the diffraction. I moved my energy towards focusing the light, but then i noticed the beam out of the collimator was not fully collimated. It was focusing at 250mm. so i shine the light across the room and reflected it back to same side of the room as the tweezer board and got the light extremely close to collimation.

Because i change the collimator, the position is no longer on the optical axis of the Objetive. Re-aligning it has proven to be difficult, but it needs to be done to continue.

497: power:770mW , Output:830mV, ECD Output: 1.26V

Got the DFG: Helpful notes: to take out the 10sec of evaporation you will need to comment the lines, in there main.pys under run sequence

      self.run(self.MOTLoadDuration+250*ms,
               Evaporate(ODTParams = self.ODTParams))

I ran out time to put in a lens and focus light onto the test target, but i can check the following off. First the waist of tweezer light was measured at 6.3307mm on the actuator.

i got the yesterday's images with only the Objective and the Tube Lens in their working distances.

I adjusted the Z-position of the test target to get the image into focus.

the 461 got into focus at 6.5420 mm

the 690 diode got into focus at 6.7357 mm

This numbers correspond to Column 5: with thicknesses of about 10um

497: power:800mW , Output:840mV, ECD Output: 1.38V

461 at focus of 6.3307mm

698 diode at focus of 6.3307mm

over the last couple of days, I proceeded to take numerous KnifeEdge measurements of the 497 tweezer. After adjusting the spatial different of the to telescope lens and checking with theory, i can confirm that the smallest waist possible is ~260nm for 497 tweezers. In order to decrease that number further, we would need over 5meters of space.

A new issue has arrived, via the beam samplers. the one currently on the experiment was too thin and causing diffraction. Grady put in a 5 mm beam, and the diffraction was gone, but now I need to realign the whole Tweezers.

The 497 lock immediately: the Output PD was 0.748V, the ECD output was 1.4V, total power just out of the laser is 730mW

After 35 mins: the Output PD was 0.83V, the ECD output was 1.1V, total power just out of the laser is 850mW

Maximum possible distance between the 497 telescope lens: Waist:268nm

about 8-10 turns closer: Waist

• Tweezers within 1 Rayleigh range and we have +/- 10um to move the 698 in the z-direction
• Fix red/blue imaging lateral offset
• Add vertical imaging (combine with the blue MOT)
• Add the MOT lens optomechanics

The 497 lock immediately: the Output PD was 0.675V(increasing), the ECD output was 1.2V(decreasing), total power just out of the laser is 625mW(and increasing)

After 35 mins: the Output PD was 0.8V(stable-ish), the ECD output was 1.1V(stable-ish), total power just out of the laser is 840mW(and increasing)

After additional 30 mins: the Output PD was 0.847V(stable-ish), the ECD output was 1.07V(stable-ish), total power just out of the laser is 910mW(stable-ish)

1/2 turn CCW: from image698_57 (17.61277+/-0.00139)mm to image 698_58 (17.61336+/- 0.00181)mm:off by 2.81um

1/2 turn CCW: from image698_58 (17.61336+/-0.00181)mm to image 698_59 (17.61500+/- 0.00173)mm:off by 1.17um

1/4 turn CCW: from image698_59 (17.61500+/-0.00173)mm to image 698_60 (17.61475+/- 0.00163)mm:off by 1.42um

1/4 turn CCW: from image698_60 (17.61475+/-0.00163)mm to image 698_61 (17.61613+/- 0.00166)mm:off by 0.04um

0 turn : from image698_61 (17.61613+/-0.00166)mm to image 698_62 (17.61+/- 0.00)mm:off by um

For the 497, starting at image 14

2 turn CW rotation of the 30mm lens (from the control) & 0 turn CCW of the 250mm lens (from the control): waist increased 0.582nm

6 turn CW rotation of the 30mm lens (from the control) & 0 turn CCW of the 250mm lens (from the control): waist increased 0.633nm

6 turn CW rotation of the 30mm lens (from the control) & 3 turn CCW of the 250mm lens (from the control): no change

10 turn CW rotation of the 30mm lens (from the control) & 8 turn CCW of the 250mm lens (from the control): no change

10 turn CW rotation of the 30mm lens (from the control) & 18 turn CCW of the 250mm lens (from the control):

The 497 lock immediately: the Output PD was 0.7V, the ECD output was 0.9V, total power just out of the laser is 890mW

The 497nm Knife Edge gave a Rayleigh range of (440 +/- 10)nm

Grady went over the types of lasers we have in the lab. We use 2 types of lasers: TiSaph Crystal and Diode.

The Diode laser use a electrical current to power the laser medium. these include 698,689,473,436,481

the TiSaph uses 1064 into a doubler to get 532, the 532 is shined onto a TiSaph Crystal which then generates light ranging from 700nm -1000nm depending how you tune it. these include 689(not on), 497(994), 461(922)

497 Tweezer, output power 0.99mW (86% eff) .115mW

i took a few knife edge measurements, but the step size were too big to get a waist, so i didnt document them

Im unsure why, but the 497 will not lock properly. I have to let the Solstice scan for the correct freq a couple of times before it will randomly lock.

When the laser locked, it locked at 0.65V ECD Output, which is half of what it was at the end of the day yesterday. I needed to recouple the wavemeter fiber and the experiment fiber. I spent some time trying to get the wavemeter eff higher, but I couldn't get past 100microWatts (which isn't enough for the wavemeter to detected).

I turned my effort to getting a higher experimental eff. It was hard to see the different beams of the AOM due to the high power, so I put some ND filters in front the AOM, and saw that I was about to run out of travel. So I unclamped the mirror, repositioned them, and started recoupling. As I began, recouple, Grady, notice that my beam size and shape was not good, and that collimator had a hard time getting a decent eff. He suggested I switch to a CFC8 collimator. After some time and Grady telling me a secret to coupling, I got eff of 50% through the experimental fiber; I used that fiber to run another Extinction ratio because of all the recoupling and got a dB of 38.

I place in the fiber that leads to the tweezer board. I noticed no space between the collimator and the PBS for the polarimator, so( with Grady's advice) I constructed a weird way to place a mirror in that small space for the light could be reflected to the polarimator. After using Grady's and mine phones for facetime, we got 45dB.

Now that we had decent power and stably polarized 497 going to the tweezer board, Grady showed how to set up the PID for intensity stabilization. I ran a BNC to SMA to BNC to the Red Patiya, at IP Address 172.16.74.xxx, which the IP Addresses labels are wrong for most of the Red Patiayas (IP Address, 172.16.74.194 is what the 497lattice PID says on the computer). When we ran into some errors, First channel 2 was not working, and we believe the fiber was coupled into the 0 mode and not the 1st order mode. I will check this next week.

I help Grady try to get a Higher OD by better aligning the Red Vertical paths and the blue mot. We determined that they were already optimized and that powers were the only thing left to optimize.

the 497 polarization is not stable, Grady suggested I couple 994 Doupler to the wavemeter to check the freq and so we can see it in the future.

I checked the 921 power into the wave meter and its only using 180 microWatts. I got 350 microwatts of 994 going to the wavemeter and I don't see anything. After some trial and error with eh 994 power and the change its wavelength. the fiber going to the wavemeter it a 780 PM fiber. THe limit for that fiber is about 965nm for the wavemeter, the manual of the wavemeter says I ned a 980 single-mode fiber.

Since I could use the 994 doubler to get a freq reading, I used the 497 instead. I was able to the wavemeter to read the 497 at 135 microWatts of power, but just barely. After I confirm the wavelength was 497, I put the experiment fiber through an extinction Ratio and got it stable as well as a >40dB

i some how was able to lock the 497 laser. I notice the power was decreasing, and relooked at the Solstis webpage. I saw the ECD ouput decreasing to 0.071V. i relocked the laser(unlock and lock the ECD), some how its now at .980V and decreasing.

.900V

0.740V after 34 min

Im testing how the power shifts as i turn the 473 HWP.

The experiment gets 35mW(11,94,192,276) to either 112mW(52), 122mW(234), 156mW(320), or 181mW(144). WATTS(ANGLE OF HWP)

The DUMP gets 35mW(55,140,236,318) to either 75-170mW({102}), 92mW(280), 122mW(9), or 156mW(189). WATTS(ANGLE OF HWP)

The 497 laser seems to be extremely senstive to touching the table. It love to come unlocked.

my working condition of 15W on the Solstis:

the total power out of the 497 laser is at 1.098W

Max power after the PBS is 636mW

Min power after the PBS is 141mW

Grady helped me troubleshoot why the laser was so senstive. He beleives that is was aligned through the double too well, and the trade off with the senstivity.

Currently, there is 120mW going into the aom. and 19mW going into the fiber. the fiber is only coupled to about 10%eff. i will get more tomorrow.

I ran the RMOT at 7pm, and the holzworth was at 329863

starting at 330425kHz

found at 329965kHz it moved -460kHz

497 Tweezer, output power 0.mW (% eff) .245mW

I noticed the power fluctuating heavily. so I wasn't able to perform any knife edge measurements. I went to do an extinction ratio but couldn't find an extra 497 Half waveplate (HWP). Grady and Khang gave me a 473 HWP to replace the 497 HWP that was being used for dumping power. and I will take that 497HWP for the Extinction Ratio.

Tomorrow, I will recouple the 497 to lab2, and perform the extinction ratio, and hopefully run a knife edge on the 497.

I tried getting the 87stir power up to 20mW, or at least 17mW(what i had yesterday). I was only able to get to 11.6mW, but grady got it to 16.5mw. That is close to the set point, but not quite there. I learn where the Blue Spectroscopy laser is, which involves the Saturation Apsortion Spectroscopy.

The RMOT was gone all day, but it was realized that the issue was with the Halzworth, it was “unplugged”. but we got the RMOT again.

I learn how to turn on the 497 Laser, and aligned it into the objective. I checked and confirmed, that its is colinear with the beam on the axis of the large objective.

HOW TO TURN ON/OFF 497:

All controls are on the Imaging Monitor

start after data:5522

085→205kHz To get a actuate drift rate for the Holzworths, Grady suggested we calibrate first. The image below shows the data of how much the RMOT moves as we change the holzworth by 20kHz. Each point is the average of 3 runs.

After we set the RMOT to a frequency corresponding to the highest OD, and let it run for about 3 hours. We used the data of the drift and calibration to get a drift rate of 1.2Hz/sec. We did this again, and we got a drift rate of 1.3kHz/sec. At least for this day, the drift rate was we consistent throughout the whole day. I will check the drift rate tomorrow, but we we only need to run for 30 minutes at a time instead of 3 hours drift1: start after 5561

331,115kHz start

drift 2: start at 6709

330,995kHz

ITS MY BIRTHDAY!

295,190kHz: TimeStamp: from 7389814716 to 7389815426

295,240kHz: TimeStamp: from 7389815451 to 738981XXXX

The RMOT moved about 50kHz from 7:30pm to 10:30am. which seems to correlate with the 2.77kHz/hr drift. I am running the experiment on loop since 10:40am at 295810kHz, to see if any significant changes.

295,810kHz: TimeStamp: from 7389804467 to 7389804871

295,780kHz: TimeStamp: from 7389804886 to 7389805150

295,700kHz: TimeStamp: from 7389805159 to 7389805376

295,660kHz: TimeStamp: from 7389805388 to 7389806145

295,510kHz: TimeStamp: from 7389806165 to 7389806430

295,460kHz: TimeStamp: from 7389806474 to 7389806787

295,380kHz: TimeStamp: from 7389806844 to 7389807113

295,330kHz: TimeStamp: from 7389807122 to 7389807426

295,270kHz: TimeStamp: from 7389807438 to 7389807742

295,200kHz: TimeStamp: from 7389807770 to 7389808496

Last center point was 295,080kHz with OD 0.377 at 7389808520

I determined the best way to align the Tweezer board MOTs were to use XY translation mount for each of the telescope lenses. THe 25” lens will be on an adpater, so it can be removed easily.

The REDMOTmight be accelerating, considering the rate of change is increasing. its currently about 1.4 to 1.6 kHz/sec.

I was only in lab for a short amount of time. I help get the red mot back. The issue seemed to be either a 2.77kHz/hr drift or a massive jump due to a magnetic next door.

I also learned more about the mot pathes of the expermient and power they should be at just before the window.

I aligned the Blue MOT, Blue Imaging, and Blue Tweezer through the objective, and they should be colinear with the red paths.

Both MOTs and Tweezers do not have their telescopes in place.

I aligned the Red MOT, Red Imaging, and Red Tweezer through the objective.

Prerequisite course work or knowledge for this project:

Basic Optics (Gaussian Optics), quantum mechanics, familiarity with electronics(microcontrollers), Python, Photonics, and Atomic Physics

Nature and frequency of contact (hours per week):

2 hours of meetings, 17 hours of lab work, 2 hours of readings

Means of Evaluation (paper, final, etc.):

• General:
• Present a Scientific Journal that related to our work
• Completion of weekly tasks
• Installing the Tweezer Board onto the main Table

• Overarching
• Trap atoms using the tweezers

• Potentially
• Measure the Magic Wavelength

Proposed plan:

I plan to start with aligning and characterizing the lasers for the tweezer board before it is moved. Tasks like making the beam colinear and the tweezers overlapping are essential. As well as imagining and camera characterization. Adding a reference laser along the axis of the main chamber's objective

Once the tweezer board is moved, I will be tasked with realigning into the main chamber, debugging, and sending the proper lasers. Then I will help with any issues that arise from running the experiment and imaging the atoms.

churn numbers and Barry curvature

I measured the angle between an alignment beam through the objective and the back couple beam on top the objective. Over a there is an angle of separation of 0.0407-0.1058 Degrees, I measured a separation of 5-13mm over a distance of 7.042meters

At the 1 meter distance the light travels on the main optics table, the distance between using the 2meters alignment techniques will be 0.71 -1.85 mm.

Finish the last HW for Phys 212B

goal of todayis to realign the 2 tweezers with the objective/

698 Tweezer:

power going into the AOD: 0.575mW

power going out of the AOD: 0.525mW Insertion loss of 8.7%

Power of the (1,1) mode of the AOD: 0.274mW (53% Eff)

698 tweezer:the 1/4 wave plate is at 248, the 1/2 wave plate is at 199

497 tweezer:the 1/4 wave plate is at 329, the 1/2 wave plate is at 331

492 diode Polarization 698 diode Polarization

We have access to AAOptoelectronic AOD/AOM Manuals. Login USERNAME: purchase2013 PASSWORD: xvd2013

http://www.aaoptoelectronic.com/wp-content/uploads/documents/DTSXY-xx-Manual-V1.pdf

Some important information inside this manual.

How collinearity work with incident beam, (1,1), and (0,0) modes

Explanation of the butterfly diffraction rings around the AOD modes

I measured the angle between an alignment beam through the objective and the back couple beam on top the objective. Over a there is an angle of separation of 0.244-0.488 Degrees, I measured a separation of 30-60mm over a distance of 7.042meters

At the 2 meter distance the light travels on the main optics table, the distance between using the 2 alignment techniques will be 8.5 -17.0 mm.

Using the near perfect alignment laser through the objective. I place the slit's kinematic mount with a mirror reflecting the beam back. I there used the tip and tilt to back reflect the light to the collimator. This tells me that the slit mount will being parallel with the objective and the tweezers.

The 492 diode fiber was broke at the beginning of the fiber, even though there were no changes to the setting and fiber,the light coming out was extremely dim. I notice very bright spot on the fiber, and that was an indication that the fiber was broken.

I did practice before but i fixed the fiber by the following steps

1. I cut off the fiber that could not be used. Then i took regular wire strippers, and stripped the outer coat off, by putting the fiber at the “scissor like” part of the strippers and rotating the fiber to get a circular cut. I was careful not to cut all the way through.

2. I placed the fiber through the temporary fiber tip with a little bit of the fiber core out of the tip.

3. I then taped the exposed tip to a table and use a optical fiber cleaver to VERY GENTLY make a cut on the fiber and then pulled it to break apart. You will know if this is done correctly if your beam out of the fiber is a Gaussian.

4. Align the end of the fiber core with the end o of the temporary fiber tip.

Be Careful.

1. Roughly Align your laser through the center of your objective. Try to get all back reflection on to the thorlabs card that you are using.

2. If possible, Place a mirror on the other side of the objective to reflection all the light back.

3. On your card, You will need to spot the brightest dot and the largest dot. The Brightest one represents the first back reflection and the largest represents the mirror in the back.You using the Near Near Far Far method to align those dots.

4. Once you believe they are align, take out the mirror and use one of the next largest dots. Repeat until all dots are over lapping. Using the more concentrated dots as near and others as far. This will take some time. It will seem like the back reflections are circling into a single point.

I plotted overlapping 492/698 tweezers: This is NOT the current alignment, just the best one.

698 Tweezer, output power 0.155mW (86% eff)

Im using tape to mark the rotation the i do to the tip/tilt kinematic slit mount.The starting position is a piece of tape on the 689 tweezer side(12oclock) and no tape pointing to the 492 tweezer(6 oclock). All rotation will be made from the looking top down view point.

Adjustment to tilt of slit mount:

Made a 1/4 CCW turn to the Y-Knob:

Unclasped the telescope lens and put them in the middle of travel: New Control:

No Changes to 698 tweezer since previous entry:

698 Tweezer, output power 0.147mW (85% eff)

Adjustment to tilt of slit mount:

Adjustment to the actuator travel:

Adjustment to the actuator travel:

698 Tweezer, output power 0.147mW (85% eff)

22 turns Clockwise rotation of the 30mm lens (from the control) & 9 turns CounterClockwise rotation of the 250mm lens (from the control):

I ran out of travel, so i moved the telescope lens. there is new definition of control

Control:Each turn is 0.3175mm.

Looks like i moved the lens too much, I adjusted the actuator travel to get a clearer waist measurement.

10 turns Clockwise rotation of the 250mm lens (from the control):

15 turns Clockwise rotation of the 250mm lens (from the control):

1 turn Counter Clockwise rotation of the 30mm lens (from the control) & 16 turns Clockwise rotation of the 250mm lens (from the control):

1.25 turn Counter Clockwise rotation of the 30mm lens (from the control) & 16 turns Clockwise rotation of the 250mm lens (from the control):

492 Tweezer, output power 0.223mW (80% eff)

Comments:

Using the last 2 images, the largest distance between the 698 waist and the 492 diode waist is ¹⁾ = 2.43microns

Using the 729-22=707nm waist for the 698 tweezer. the Rayleigh Range becomes 2.249 microns. Giving us a possible distance of 180nm remaining.

698 Tweezer, output power 0.147mW (85% eff)

The steps seems too erratic,

i change 40 slices to 100 slices

I increased the travel, did 50 slices, each slice was not completely finished. Also, I rechecked the alignment into the objective,

best data yet, moving onto 492

492 Tweezer, output power 0.223mW (80% eff)

I increased the slices and data amount

698 Tweezer, output power 0.147mW (85% eff) adjustment to the focus of the 2 telescope lens.

Control:Each turn is 0.3175mm.

1 turn Clockwise rotation of the 30mm lens (from the control):

2 turn Clockwise rotation of the 30mm lens (from the control):

3 turn Clockwise rotation of the 30mm lens (from the control):

1 turn CounterClockwise rotation of the 30mm lens (from the control):

1.5 turn CounterClockwise rotation of the 30mm lens (from the control):

20 turn Clockwise rotation of the 30mm lens (from the control):

Comments:

the 20turns ~6mm travel. The waist increases 70nm to total of 650nm. with a 50micron focal shift. We need a 90micron shift remaining.

If the expected atom spacing is 630nm, the Rayleigh Range of 698 would be 1700nm. for 2 atoms, spacing of 1260nm, the Rayleigh range would be 7100nm.

The 698/689 bandpass filter has a reflection of 94% efficiency

The input 492 tweezer power is 0.185mW

698 Tweezer, output power 0.134mW (84% eff)

Comments:

I suspect the slit is tilted in some way. Giving the measured waist slightly off from the fit waist

focal_shift.pdf

I switch the 698 AOD for the proper one. The power going in was 566mW. output power of 523mW for efficiency of Insertion loss of ~7%. The Beam size is equal to the beam waist because its a columnated beam, which is 1.4mm. the aperture of the 689 AOD is 7mm.

The (1,1) mode has an efficiency of 56% at 290mW

This is the 492diode tweezer, measured through the same set up as the 698 tweezer, the graph axis should be comparable to get the focal shift between the 2 tweezers.

Side Note: Tyler help me relearn how intensity stabilization works.

1st: you add a beam splitter to you path, and send the light into a photo diode.

2nd: the photo diode will send any changes of intensity to a Red Pitaya

3rd: the Red Pitaya will control a AOM through RF, to change the laser .

When i took off the the tube lens, to switch for the 2x 30mm lens, i re-measured the 492nm input power. Turns out i was getting 83% efficiency the whole time. I had incorrectly measured the input power.

Using the 2x 30mmlens, the 492 output power was 64% efficiency.

I went back too the original lens tube setup, and i got the 698 efficiency to 88%

492 Diode waist, Rayleigh Range of 399nm

497 waist, Rayleigh Range of 258nm

698 waist, Rayleigh Range of 1675nm

Finding when the waist become 1/3 the len diameter

Len's Formula for Gaussian beam (698nm, after objective)

Placement for 1000mm lens

Where 698 focal shift is 240mm, the 1000mmlensshould be place -1.051meters before the tweezer

using the waist equation: Tyler saidthe698nm laser has a 11mm waist, after the 1000mm lens it becomes 20.2microns, after the objective lens the new waist becomes 20.1microns at s=0, and 173microns at s=16mm.

Table: This shows how the tweezers diverge. The results are the distance when the waist is 1/3 the lens diameter.

The ω_o used with 250nm for 492, 610nm for 698, and 202nm for 497

492 Diode, with (4.5/30/25)lens tube, max output power 155mW (83% eff)

1st Attempt

1st Attempt

698 Diode, with (4.5/30/25)lens tube, max output power 0.293mW (88% eff)

1st Attempt

1st Attempt

2nd Attempt

3rd Attempt

I coupled 698 laser to lab 1. The output power is 0.8mW (10% eff).

I noticed the x-AOD would fluctuate as I moved the SMA wire.

I determined the 698X SMA wire was broken,so I replaced it, and it now shows no issues.

The (1,1) mode out of the 698AOD has an output power of 0.43mW (53% eff).

The 698nm tweezer has been aligned through the objective, with the 1000m lens.

I optimized the RF Amp connectors to just be BNCs,so we dont have to worry about the SMA connectors. I label the RF amps to 698Y, 698X, 497Y, and 497X.

RF AMP:Output (dBm):Output PWR(W):

698Y:31.1dBm:1.29W,

698X:31.2dBm:1.32W,

497Y:28.4dBm:0.69W

497X:28.4dBm:0.69W

I check to see if all the RF AMPs are working.

I learned to better navigate the tweezergrid.cpp. Grady suggested a single file is made that works, and I will use a copy to make mistakes on. The 698 AOD has an internally frequency of 122Hz and the 497 AOD is 95Hz.

I made a README.txt has the instructions to get the AODs working when you change between each frequency.

Julio and I determined the 698 AOD still works, We believe the SMA connector to the RF Amp was loose. Should be aligned into the objective this week.

this 100 data points for 50 slices

Input to termial for 180 X 100 { 26003909 0.000471 180 26003984 0.000502 100 }

150 points, 1 slice

180 points, 100 slices, Tylers code: 2.2 micron waist

180 points, 100 slices, Gradys code : 220nm wasit

I determined a good enough combination of lens is the flat side of 4.5mm lens, curve side of 30mm lens, and finally the curve flat of 25mm lens. With this combination, a get about 63% eff through the slit and lens tube.

The image below is my first semi-descent waist for the 492 diode laser

I spent the day combining a series of lens that will focus the tweezers on the power meter and keep majority of the efficiency. I determined the best way to do that is to use 3 Achromatic lens inside a lens tube. Because of how divergent the tweezers are, the 1st lens will need to have a small focal length. I use 4.5mm lens. Im unsure which lens to next, I currently have 25mm, 2x 30mm, and 60mm lens

Tyler and I got the Frame built.

Using the Dell Laptop

File name: AutomatedKnifeEdge.sln

In Andaconda3 folder, Open Andaconda.Navigator, then click CMD.exe Prompt

in the command line go the this dir: C:\C#Programs\AutomatedKnifeEdge\AutomatedKnifeEdge\bin\Debug

run this code: AutomatedKnifeEdge.exe FirstKSTSerialNumber(Horizontal) [jog increment (mm)] [# of steps] SecondKSTSerialNumber(Vertical) [jog increment (mm)] [# of steps]

How the Actuators move with the inputs: the KSI_1 will move the (jog.Incr)x(steps) recording data after the first movement. The KSI_2 will do the same, but will recorded data before the intial jog

it will create a csv file located: C:\Users\DETECT\Documents\KnifeEdgeData

There are 3 python scripts that you will use to plot then fit the data. Each will be ran in Spyder. You will have to change the the file name within the code itself. You should only move onto the next script when you see the idea plot in the 1st python script.

Open Spyder through Andaconda.Navigator

1st Python Script: PlotAndFit.py Located: C:\Users\DETECT\Documents\KnifeEdgeData You should see Image 1

2nd Python Script: FitWaist.py Located: C:\Users\DETECT\Documents\KnifeEdgeData You should see Image 2

3rd Python Script: PlotAndFitBothSidesMultipleSlices.py Located: C:\Users\DETECT\Documents\KnifeEdgeData You should see Image 3

Image 1

Image 2

Image 3, this is what the data should look like.

Where you get the Waist

The was my first Try.

I aligned the 497 laser to go through the objective. Troubleshot why the AOD was producing a Multi-Tone (striped) image, with Robin.

I recoupled the 698 laser to feed through to the tweezer board.

I found the proper part and placements for the safety rope,, that we will use for transporting the tweezer board.

We will use 4 eyebolts per corner of the board, 32 ft of rope per corner, and 4 climbing ascenders.

Waist Slit Help: Tylers notebooks from 9/8/2021 to 11/1/2021

https://wiki.barreiro.ucsd.edu/lib/exe/fetch.php?media=sr:thesis_ivaylo_madjarov.pdf

reset interlock warm-up start, Grey dot on the right turns green dot below tells you about the shutter Open or closed

to turn off press idle then close shutter

user:pass::main:main

to optimized align: press One Shot

Holzworth 00.329.283.000.000 → 00.329.288.000.000

image is from the 497nm diode laser

Tyler and I were able to the 497 light into the Main Objective and focus the light to the focal point of the lens.

We add the test target and reflected the light across the room and across again determine the how large the Focal shift is. Which is, when readout by the actuator, 1.9093mm to 1.9365mm (27.2 micron focal shift).

i sent the 80/20 pieces to the machine shop, they said it should be done by tomorrow.

I installed the picomotors.

Here is the manual for the Picomotors

Here is the manual for the controller.

At a quick glance:

The motor has a 12.7mm travel length. with a 52.9nm/per count. The total count is 240076.

Labels

Controller 12127 Motor 1: Motor 1

Controller 12127 Motor 2: Motor 2

Controller 12030 Motor 1: Motor 3

All Motors are currently Zero'd out at their center of travel.

Motor 1 can travel ± 31200 counts (or ± 1.65mm)

Motor 2 can travel ± 52000 counts (or ± 2.75mm)

Motor 3 can travel ± 33100 counts (or ± 1.75mm)

Holzworth 00,329,240,000,000 → 00,329,242,000,000

This is the center of the star, with the 497nm light in the blue image path.

The Acquistion files are in strontium/equipment/Andorcamera

I was able to capture an image of the target with the blue path. The significant changes i made to the board was that I added in an iris directly after the flashlight and adjusted the kinematic mirror that guided the light into the Lrg objective lens.

There is a large difference in the magnification of between the red path and the blue. I believe this is due to the light diverging. The Blue path is about 14“ longer then the red.

Blue path:

The blue light is very dim, so the acquisition setup is: 100kHz readout on the Conventional Amp with a exposure time of 1sec.

The focal depth from the target was 14.0306mm in the actuator, and at a horizontal position of 3.6112mm on the other actuator.

Magnification~ x616

Red path:

The acquisition setup is: 10MHz readout on the EM Amp with a exposure time of 0.006sec.

The focal depth from the target was 14.4393mm in the actuator, and at a horizontal position of 3.7385mm on the other actuator.

Magnification~ x172

Holzworth 00,329,232,000,000 → 00,329,235,000,000

OD is about 1.00 after 2.5 hours

The red imaging path still shows the target, while the blue one does not. I measured the path difference, and the red path is approximately 14” shorter the blue path. This could have an effect on the depth needed for the target.

Grady advised to check the focal shift of the different lens of the 2 telescopes. That could have something to do with it the focus of the light.

Main Objective Lens has a focal shift of 0.6mm reference: Strontium/construction/objectives/Purchased objective/na08

75mm Lens has focal shift of about 0.3mm reference

Imaging Lens has a focal shift of (thorlabs says it negatable) reference

Tube Lens has a focal shift of 0.05mm Reference

Imaging of the blue path. Edge of window can be seen

With a quite a bit of help I got the BeagleBone Black to work. I had to use the Dabien firmware and not the eMMC firmware. It took a while, but it worked on mac.

I checked the alignment of the blue image path, and its still aligned. The camera still sees the target from the red image path.

When shining a white light through the blue image path. We see a blue light going in the image objective lens, but almost no light out of it. Robin recommended i use a power meter to detect the light going into the camera. The power after the image objective lens drops from 26nW to 1.6nW

I took off the imaging objective lens to see if the light was aligned with the z-axis. It was not aligned. I proper aligned it with the red diode laser, the 2 kinetic mirrors and the last dichroic mirror. After the alignment, i was able to see the light source on the camera (which was new), but not the target images.

We changed some setting on the camera, We switched from EM Amp to CCD Amp, becuase it allowed for 100kHz readout. And then we adjust the exposure to the highest value of 9 seconds. This was an improvement because we could see the edges of the window lens, but still not the target.

After inspecting the window/target/objective area, i noticed the window moved to a lower position then it usually is. I moved it back, which took the window out of focus, but the target wasnt seen to be in focus.

The original focus position for the red imaging path was 14.4133mm on the actuator.

All and all, the blue imaging path seems to be aligned because we can see the light source and window, but i still need to put the target into focus, the only issue is the 9 seconds long exposure time.

We plan to transport the tweezerboard to a cart then to the new table. I designed a frame for the cart for the board.

I drew out the lines on the 80-20s to cut them, plus labeled each part.

I aligned the blue image path but currently cant see an image on the camera. Its hard to see the blue light because its too close to the edge of the visible spectrum. I hope the camera will be able to see it.

Current Sketch of the Tweezer Board Frame

Cool image of the white light and it reflecting the different colors

Counted all the 80-20s that we had and measured their lengths. We have enough to build the frame.

I then work on aligning the Blue Image path

Briefly talk to Tyler about the frame design

I drew out a rough design of the Tweezerboard frame. It seems to be a good working idea, but it needs to look over with Julio.

Not much work was done. I just roughly setup the diode laser and the 461nm path for the follow day.

I spent a couple hour trying to get a higher 461nm efficiency to the tweezer table. Its currently at 0.1mA (11%). Tyler suggested i move on to the imaging, because it is more important.

I did get a lot of coupling experience.

I have been fiber coupling the 461nm Laser to tweezer board. I have a 43% efficiency going to the table, but im having trouble getting a high efficiency to the optical equipment.

Here is the infomation needed to determine the magnification with the IXON Ultra Camera.

R1L1S1P Target

Thickness of each lineUSAF 1951.

The iXON has 13um pixel width and other hardware info can be found Here.

I got more practice fiber coupling and increasing power efficiency of blue laser to the tweezer board.

Holzworth 00,329,150,000,000 → 00,329,155,000,000

195X

After adding the 2nd telescope, we lose some clarity in the image, but not too bad. I believe this clarity can be optimized, but i just ran out of time. The Theoretical magnification is 187X.

Holzworth 00,329,145,000,000 → 00,329,150,000,000

Tube Lens 180A, the tube len's working distance is 133.5mm and the focal depth of the ixon is 17.5mm, so the distance between the tube lens and camera should be 116mm apart.

Without the Window, the Working distance for the focal length of the tweezers objective lens in about 7.5 to 8mm. I determined this by setting the tube lens to the 116mm working distance, finding the most focus point on the target looking through the camera, and confirmed the magnification to be 11.6X to 12X.

Holzworth 00,329,130,000,000 → 00,329,145,000,000

Sedlik2022