MAPS-2024B – Extreme Wavefront Control Lab

MAPS Aug. 2024B Night 6: On and Off

Welcome to the last blog for the MAPS August 2024 observing run!

The evening started off with Grant and Oli performing an LED strip test that would allow us to more clearly see out of focus images on our pupil lens and visible wavefront sensor. The LED light strip was placed in front of the ASM, along the radius of the shell. Our pupil image was clearly out of focus. Ideally, the circles of light (each individual LED bulb) should not be overlapping. They should look as they would if you were to look directly at the LED strip itself. The VIS WFS image looks better, as you can see more defined pixels of light.

Out of focus pupil image. If you own an LED light strip and ASM, you can replicate this at home!

LED strip test on the VIS WFS. WFS pupils should not normally look like this.

After waiting for a few storms to dissipate, we get straight to doing some AO. Amali closes the loop on 50 modes, then closes on a whopping 100 modes shortly after. The 50 mode loop had a 1.4 lambda/D full width half maximum, and the 100 mode loop had a 1.8 lambda/D FWHM. GO TEAM! Tonight’s honorable target was FK5 0672, theta Herculis.

Open loop for comparison. Pupil and PSF quality courtesy of atmospheric turbulence.

50-mode loop on and off. Video taken by Amali Vaz.

Bonus Jared quote of the night: “We need more monitors.”

Now we hand things over to Grant and Oli. They battle the on and off cloudy weather as they work on VIS pyramid alignment.

Starlight, no starlight, starlight, no starlight, starlight, no sta-

After they scrambled the topbox (moved the vis CTL and CCID75), the VIS WFS pupils looked VISibly better.

After topbox alignment. The clouds took away some starlight on our pupil image.

With our new, beautiful WFS pupils, Amali, Bianca, and I take turns taking 50 and 100 mode response matrices. The image below is a 50-mode self RM, which confirms we are sensing the same modes that are being poked.

Unfortunately, AO had to come to a halt with about an hour left of observing time. While taking some RM’s, we noticed a giant red blob on our actuator position and current map. Red blobs are not good, as they indicate contamination in the ASM. What could this contaminant be? We are not sure yet, but it could be anything from a detached magnet to a bug seeking warmth.

That brings us to the end of the August 2024 MAPS run! I haven’t even hit the one year mark of working with the MAPS team, but I can still say that I am proud of how far we’ve come since my first run.

Goodnight MMT, goodnight air, goodnight stars, goodnight ladybugs and abnormally large daddy long leg spiders that are everywhere.

:･ﾟ✧:･.☽˚｡･ﾟ✧:･.:

As we all make our way home, I would like us to keep the song of the night in mind: “On the Road Again”, but specifically Donkey’s version from Shrek.

MAPS Aug. 2024B Night 5: Simulations & Storms

The duality of nature is something I’ll never stop wondering at: my first post came to you on a night so action-packed, I felt incapable of writing every detail down. Last night, however, we didn’t leave lightning shutdown.

To bide the time in hopes of some sky-time, Amali guided me through the MAPS CACAO simulation; from starting up the server, deformable mirror and wavefront sensor, to closing the loop with 20 modes, 50 modes, and finally 100 modes. Hopefully this means we’ll be able to reach 100 modes on-sky!

MAPS CAAO Simulation: 20 Modes!

Next, Lauren was shown the ways of simulated CACAO. It was comforting to know we wouldn’t break the ASM since this was only a simulation, and it was alright to push the mock-ASM to its limits and discover what not to do to the real deal. Apparently, completion of the simulation means I am now prepared to operate the AO system on-sky solo. But, I still feel like a tenuous sapling whose very existence/job/ASM is threatened by anything stronger than a mild breeze.

Amali’s pedagogy then dictated Lauren and I teach Parker how to operate the MAPS simulation. Thusly, we discovered the gaps in our learning and may have unintentionally confused Parker in the process but, he did succeed in reproducing a 20-mode loop. Maybe we didn’t do so bad after all? Our next step towards becoming strong AOaks is to write the simulation steps in our own words. Gee, it’s a good thing lab classes were a requirement for graduation; who would have thought writing reports was so important? (I jest, I jest)

The word of the night is Arizona. Talking to Yoav, I was reminded that not only does Steward Observatory and the University of Arizona contribute significantly to astronomy and humankind’s exploration of space; but so does every University in Arizona. We are strengthened by our inter-university collaboration, now more than ever. In this sense, I dare to say Arizona means ad astra per intellectum et cooperante–to the stars through understanding and collaboration.

Now, I would love to explain the majesty of the storm that kept us from opening, but words would not do it justice. So I’ll let the photography of our talented telescope operator, Brian Pinault, speak for me. Ever the diligent guardian of our invaluable tools, Brian captured this moment of volatile yet sublime nature while visually surveying the storm’s trajectory.

In honor of the storm, the song of the night is The Tempest, Op 18 by Pyotr Ilyich Tchaikovsky.

I’d like to add a bonus to tonight’s post. If you don’t already listen to Ologies, the science podcast by Alie Ward, then you are missing out. Let me present to you one of you new favorite listening experiences with the episode on Fulminology, the Study of Lightning. Check it out, or your chances of being struck by lightning may increase! (again I jest, I jest)

Maybe devil fruits don’t exist in our universe, but lightning does…

MAPS Aug. 2024B Night 4: 100 Modes and Tornadoes

Hello! My name is Lauren. Similarly to Bianca (introduced in the last blog post), I am research staff for CAAO, and the second padawan under the great Amali Vaz. This is my sixth MAPS observing run, and I have been given the duty of making my first blog post! Since we are currently clouded over, I will mostly be filling you in on the non-AO happenings of tonight…

As soon as I was awoken from my slumber by the sound of thunder, I knew the day that we had all been dreading (just for astronomical reasons) had arrived. Once we got to the summit, I decided to enjoy the cloudy weather and see what the great outdoors had to offer today. Out of the corner of my eyes, I saw two little blobs hopping around. Coatis!

This coati and I had a staredown for a minute or so.

The second coati quite literally huffed at me and ran away. I was a little offended that the wild animal did not sense my calm, animal-loving spirit.

Besides rain and hail, there were other interesting meteorological happenings in the area today. First, a tornado warning was issued early this evening in southern Pima County. Just a couple hours later, another tornado warning occurred in Dragoon, AZ. This led to the important conversation of where we should go for shelter in the case that a tornado rudely passes through the MMT.

Pop quiz:

If a tornado warning occurs in the MMT area, what should you do?

a) Seek shelter at the top of the dome b) Seek shelter in an area away from windows and external walls c) Rotate the building at the same speed and in the opposite direction of the tornado in order to counteract the tornado (Yoav’s idea)

Artist interpretation of Yoav’s idea. MMT image courtesy of Frank Taylor

One good thing about cloudy nights is that we can always work on software. Thanks to Andrew and Orlando, we can now control filter wheel positions through INDI, which can also be selected in the INDI GUI.

In addition, Amali and Jared set up the wavefront sensor simulation with CACAO! They successfully closed the loop with 50 modes.

Alas! The sky has cleared and now we can do some AO. Amali gets to CACAOing, working on taking 100-mode response matrices, while closing the loop on 50 modes. While doing so, we noticed a nice core and parts of Airy rings on the MIRAC PSF. The honorable target was FK5 0857, eta Pegasi.

Midway through this, a minuscule cloud perfectly obscured our target.

However, all good things must come to an end. There are storms developing nearby, and we close for the rest of the night.

The song of the night is “You Spin Me Round (Like a Record)” in honor of the rotating MMT building.

MAPS Aug. 2024B Night 3: Temperamental Temperatures

Howdy! I’m Bianca; formally I am research staff for CAAO, but you may have heard of me as one of the padawans under the tutelage of AO Jedi Knight Amali Vaz.

This is my 5th MAPS observing run and my first blog post! Last night was the most action-packed night I’ve yet to experience, so let’s start at the beginning.

This night we were joined by Jared Males and first-year grad student, Parker.
6 pm found us preparing for the night’s activities. The clouds got the opening time memo a few hours later, but they couldn’t stop us from being productive!

Andrew bestowed upon us Version 2.0.5 of mapspyindi2, which fixes crashing GUIs and optimizes setting the state of the ASM.

Amali and Lauren (my fellow padawan) began latency measurements round 2 electric boogaloo with new tuning parameters, courtesy of Jess.

Jared gave Parker the grand tour of the MMT and the run-down on MAPS.

Yoav hit a snowbank with the MIRAC detector: it was significantly below the recommended operating temperature due to the previous night’s efforts to cool the detector followed by a day without the heater on. Yoav led Lauren and I in detector-warming efforts for ~ 3 hours, estimating it would take 2+ days to reach our necessary temperature unless we could raise the power being sent to the heater by ~100%. Proving his multi-faceted savvy, Yoav introduced those of us less versed in memes to our word of the day- Cowabummer.

I’ll let this meme demonstrate it’s proper usage:

Original image from the 2015 DC Comic Batman/Teenage Mutant Ninja Turtles #4. Dialogue adapted and meme posted to Facebook in 2020 by user Raven Perez. Referenced by Yoav Rotman, which inspired yours truly to further adapt the meme last night.

Jarron saved the science day by instructing us to bypass the PID loop and send “UNLIMITED POWER!!!” (well, 100% but I’ve heard it both ways) to the heater.

Can you blame me? Of course I’m taking the low hanging fruit

Once the detector reached the optimal temperature, Yoav came in with the assist for Mara, a fellow ASU grad student, by taking data she needs for her thesis. Later, a closed-loop 50-mode K-band MIRAC PSF was also attained!

MIRAC PSF in the K band with AO loop closed!

MIRAC-BLINC temperature crisis resolved and the clouds having sufficiently departed, we switched gears to alignment and pinned down the order in which to move the 2 periscopes and move the telescope to have the star simultaneously in focus on the acquisition camera, entirely visible on the pupil imager with nominal vignetting, and centered on the pyramid. After the previous runs’ hard work and inevitable frustration that alignment was never quite right, the successful alignment was a small but significant accomplishment.

Pre-alignment pupil left, aligned pupil right: bye-bye vignette!

CACAO time! We set to collecting as many 50-mode response matrices as possible but temperature issues now decided to make the ASM its new target: and boy did she feel the heat. We alternated between taking 2 response matrices, followed by cooling the ASM for ~10 minutes until the sun commanded we cease.

The song of the night is ACDC’s “You Shook Me All Night Long”, because that’s exactly what we did to the ASM.

MAPS Aug. 2024B Night 2: Unstable

Tonight we first took a bunch of ASM latency measurements with different tuning parameters for Jess — took about 3 hours to get 5 iterations, may do more tomorrow night. (Well, this was after being closed for clouds for the first couple hours of the night, same as last night.)

Then we moved on to CACAO calibrating and testing. Amali tried closing the loop using the response matrices from last night. The 20 modes one worked OK, but the 50 modes one was quite unstable even at medium modes (10 and up). So then we tried median combining them and looking at them here. In the following plots we have 5 iterations of the 50-modes response matrix, and the 6th frame is the median. Tip looks OK, tilt looks OK, and focus looks OK. But they also have odd higher-order signals that AREN’T averaging out. Like the bright white or black volcano. However, these 50 modes were taken with the 20 modes loop closed, and what if that was introducing some noise? So now we are trying going back to the 20 modes response matrix and doing more iterations to average those. For the 20 modes response we are closing the loop on tip/tilt at 0.30 gain and on focus at 0.15 gain. We’ll just keep doing these until sunrise, then average them and try them tomorrow night.

Oh and Yoav had to run up the hill away from a bear as he was coming up to the dome from the Bowl tonight.

The song of the night is “Stressed Out” (Acoustic version) by Twenty One Pilots: