Update from the South Pole

It’s been a couple of weeks since our last update and they’ve been busy ones. As of our last update, we’d just started to cool down the BA1 cryostat for the first time at the South Pole. After about 2 weeks of cooling down to its 4K base temperature, we were finally able to cycle our 3 stage Helium Fridge and reach the sub-Kelvin temperatures required by our detectors.

In the weeks that followed, we undertook a calibration campaign to characterize the properties of our detectors and the receiver, some of which can be substantially different at the South Pole where we are shielded from almost all wireless communication and looking at cold sky rather than a room temperature ceiling. For a few weeks, the receiver team was kept busy collecting and analyzing data in order to figure out how well the actual receiver measured up to its design, and how to set things up for optimal operation.


In the meantime, the mount team was also busy. As of the last update, we finished construction on the Bicep Array mount and enclosed it in insulating fabric and foam panels. Once enclosed, we were able to work up inside the mount without all of our bulky ECW (Extreme Cold Weather gear).


After a few weeks of work re-installing the subsystems which had been integrated in Minnesota, we began to hoist receivers into the mount. Although Bicep Array will eventually hold four receivers, we are only deploying a single one for the first year. The remaining three slots will be filled with Keck receivers from the experiment that Bicep Array is replacing. We already know these receivers very well, so we were able to begin hoisting them into the mount as soon as we were ready.


We’ve also spent some time getting our star cameras up and running. Each of the Bicep Array receivers has an independent optical camera attached to it. Every two weeks we take time out of our normal observations and point the telescope at a series of stars. Ideally these stars are centered in all of our star cameras, but in reality the receivers and the harnesses attaching them to the mount will flex as everything moves. This flexure will move the stars slightly off center in the optical cameras and the offset will depend on the orientation of the telescope. We can use these offsets to refine our model of the true pointing of the telescope and track them over time to see how things settle. Setting up these star cameras is no easy matter however. There is no way to know just how closely they’ve been aligned when they are first installed, or whether they are focused correctly. After a long time staring at a gray screen while making very small movements with the mount, we finally saw a vague blur that looked real. And after some focusing adjustments, we saw our first star.



After our on-the-ground calibration of the receiver had finished, it was time for it to be installed in the mount. It was a big moment for everyone here, and no one wanted to be left out. Which was good news, since its heavy, and required several people to push it across the room and up the corridor to the mount. After a few hours of effort the BA1 receiver had been installed and all slots in the mount had been filled. Although it was quite roomy at first, it became significantly harder to move around inside the mount with four receivers and all the electronics installed.


With the cryostats all installed and cold, we had one final challenge ahead of us. Get real, useful data from all the cryostats at the same time. A slightly complex affair that required figuring out how to get each cryostat and its data flow working and stable in the new system. Just a few days ago we were rewarded with time-streams,  BICEP Array was officially observing.



We still have about two weeks until the station closes for the winter and most of us have to leave and there’s still a fair bit of work to be done during that time. But with a “first light” of sorts, we took time out to take a group photo in front of the new telescope with some of the amazing people that have helped to bring this project together.











BICEP Array Coming Together at the South Pole

By Jamie Cheshire:

BICEP Array deployment is now well underway at the Martin A. Pomerantz Observatory at the Amundsen-Scott South Pole Station. Before work could begin on assembling the mount, the on-site team had to decommission the existing instrument, the Keck Array, and remove the old instrument’s mount (which was originally installed for the DASI experiment twenty years ago!). After running some last-minute diagnostics, the five cryogenically cooled Keck Array receivers were unloaded from the mount and warmed up in the lab. Three of these will run inside the BICEP Array mount this first year (2020), while the other two will be packed up and sent back to North America.

After the weather cleared, the planes were able to get in, and all of the mount pieces arrived at the Pole. The crane is only able to operate during the very warmest part of the season at the South Pole, so the pressure was on to get the remnants of the DASI mount out and to begin installing the BICEP Array mount as soon as possible.


As of writing, the new base (the “triangle”) has been installed on MAPO’s roof, and the mount has been built up to the theta stage. Work is currently underway on installing the insulating fabric boot, as well drive systems and other electronics. This must be done very carefully to avoid exposing delicate cables and electronics to frigid Antarctic temperatures.

In parallel, the receiver team has been hard at work getting the 30/40 GHz “BA1” receiver ready to go. A heroic, down-to-the-wire effort from the JPL/Caltech team meant that we were able to populate BA1 with a full complement of 12 cutting-edge detector tiles for its first season. The first cooldown of BA1 at the Pole will begin imminently, after which the team will begin instrument characterization and calibration measurements.

Trucks Roll Out

It’s been a busy two months for us here.  After putting up the last post, our next task was to take everything apart and get it shipped out. So, over the course of about three weeks we dismantled everything with careful labeling and packed it away. In the end the vast majority of hardware went on one of two trucks, similarly to how it arrived. A single flatbed truck took a forebaffle and the three main ring segments, while the rest we loaded into a shipping container. These trucks made their way to California where they were offloaded for ocean shipment to New Zealand. From there, they will be repackaged onto air force pallets and fly on a C-17 to McMurdo, then a C-130 to the South Pole.


Right now we’re just at the beginning of deployment season, and the first members of our collaboration are on their way South. In the next two weeks I’ll hopefully be writing these posts from the South Pole and continuing to update on the progress of the experiment.







The Front End

With Fall rolling in, our available time has been rapidly running out. In order to ship to the South Pole, everything must be packed and shipped out of Minnesota by the end of September. Giving all of September over to disassembly and packing meant that August was a month of long weeks as we tried to finish everything that we needed to.


Since we’re in the shipping stage, I’ll be spacing out a couple posts that showcase what we’ve accomplished in the last few weeks.


One item on the list was the forebaffle; so-called because it rides at the front of the receiver and blocks light. Each of the Bicep Array receivers will have one of these large cup-shaped pieces that rides along with it in the mount. Its two halves are rigidly attached to the structure, so from the point of view of the receiver it doesn’t ever move.


The upper lip of the forebaffle just clears the main beam of the receiver, and the inside is coated with a broad band absorbing foam. What this means is that any secondary pickup from outside of our main beam must reflect first off of the absorbing forebaffle which significantly lowers the amplitude of any reflected signal. Two particular sources of potential pickup that the forebaffle intercepts are from the ground and the groundshield. The groundshield (shown below) is a large screen that prevents the telescope from directly seeing the ground which is much warmer than the sky. However, the groundshield itself is also a warmer object and light it reflects can be polarized and should be blocked from directly entering the telescope. Light from the ground can also diffract over the edge of the shield towards the receiver. The forebaffle intercepts these rays and helps to dampen any response to them that the telescope might have.



Having such a large cup-shaped piece presents its own complications. The absorbing foam has to be protected from the weather, and we can’t capture snow inside it since water (and ice) scatters the very light we want to observe. For the first issue, after we cover the inside with the absorbing foam, we add on a thin protective (and smooth) layer on top to keep ice from catching in the porous surface of the absorbing layer. The second issue is solved by wrapping the exterior of the forebaffle in a long winding of heat tape, then multiple insulation layers. During operation, we heat the forebaffle to just a few degrees above the ambient temperature which helps keep ice from sticking to even the smooth surface of the protective layer.


One other item that’s shown in some of the above pictures but that I’ve neglected to talk about so far are the insulation panels.

Many of the photos I’ve been posting are unique, in that the inside of the mount is visible from the outside. However, the vast majority of the structure will poke out of the building where it will be installed. Since we have a lot of electronics and other equipment that ride along inside the blue steel, we can’t let it be open to the cold air. To help fix this, the mount has insulation panels for its top-most and bottom-most faces. These panels are foam-core with an Aluminum skin which keeps them light but effective. In order to make them fit as well as they possibly could, they came without any holes drilled in them. Instead we had to arrange them in-place with clamps then mark and drill the holes before attaching. The top section had one additional complication, the central circular pieces need to be able to rotate without scraping along the octagonal ones just outside, and by their nature the panels aren’t especially flat. After a few days of inserting and removing shims, we were able to get everything to move quite nicely.



I’ll finish here with another video. We’ve tested the stability of our control system a few times, below is an excerpt from one of these tests. A short video of the mount scanning under computer control. It was left this way for about 48 hours which is the length of one of our typical scanning schedules. The motion you see here is typical of what it will be doing constantly for the next few years.

VID_20190816_233013 from Mike Crumrine on Vimeo.





Automated Star Pointing

By Grantland Hall:

Over the past two years, I have been working diligently on the control system for BICEP Array. That encompasses all of the software that is responsible for moving the telescope, capturing data from the receivers and other sources, and then pulling all of that data together and storing it onto hard drives. Usually that work doesn’t leave too much to look at. Even our user interface, the Generic Control Program (GCP), looks superficially unchanged, even though the underlying code base has seen over a million lines of code changes!

While our telescopes largely run themselves during winter observations, there remains one major manual task that the winterover retains manual control over – optical star pointing. This is when we use a small optical camera, pointed in the same direction as the telescope, to find stars in the night sky, and then use those known star positions to calibrate the absolute pointing of the telescope. This task is typically performed every two weeks and takes about three hours diligently clicking on images of stars. Sometimes, due to the buildings slowly sinking into the snow, the telescope mounts need to be re-leveled, causing the pointing to change, and so a full 7 hour shift of star clicking is required.

Of course, this is not an insurmountable challenge for a modern computer. So a few weeks back, after becoming frustrated with one of the electronic components needed to continue to run Keck receivers in the new BICEP Array mount, I decided to spend the afternoon trying to automate the process of star pointing. Out winterover engineer Nathan had already purchased some cameras to be used for optical star pointing, so I stuck one to the top of the telescope with some foam tape and a fisheye lens and set to work writing the software necessary to automatically locate stars.


Nathan made an offhand suggestion about dangling a headlamp for use as an artificial star, so I wrapped a headlamp around the overhead bridge crane in our highbay to allow me to dangle it over the telescope and have the telescope track the headlamp around the room. It turned out that I was restricted to a small section of the highbay due to the sun beaming in the highbay windows and drowning out the light from the headlamp.

MVIMG_20190802_162743Finally, after about a day of figuring out how to produce usable images from the camera, how to translate those images into motions, and how to communicate those motions back to GCP, I ran our very first automated star pointing! Because the fisheye lens severely distorts the image near the edges of the frame, the movements weren’t very precise, which should improve when we switch to our rectilinear lens (a 500mm cassegrain lens, for any photographers). But I was quite happy with the results!