« The Routine | Main | The many faces of eating at CERN »

Knowing the hardware


Tingting in the testbeam area.

Most of the days at CERN are spent in front of a computer, meeting over coffee, or presenting talks. We manipulate data, we talk about calibrations, we discuss quantum mechanics, and we ponder the event displays which show beautiful cascades of particles. It's easy to forget that everything in the ATLAS experiment relies on the hardware of the LHC and the ALTAS detector. So when Tingting, an SMU grad student, asked a question about the detector I decided it was time to arrange a trip to see some equipment!

The question was simple, and the answer was straightforward, but it was only obvious if you've seen detector components in person. Tingting asked why there were large "gaps" in the liquid argon calorimeter. Surely nobody would design a particle detector with gaps in it unless they had to. The gaps were a solution to one of the biggest problems of detector design- how to get all the information out of the detector. The ATLAS detector is huge, and it's really a series of smaller detectors, nested one within another like a set of Russian dolls. Somehow we need to get the data from the inner parts of the detector to the outside, and that means stringing hundreds of meters of cables through from the center to the outside. This is a very complicated task and you can get dizzy following just one cable. To make matters worse, the beams have their own equipment, including huge magnets and the beampipe, which take up space. Add the enormous toroidal superconducting magnet and pretty soon you have a very claustrophobic detector.

Of course, if you've never seen a beamline, and never seen detector hardware, then it's very difficult to appreciate these problems. Over two days, Tinging, Prof Stroynowski, Prof Sekula and I visited different sites across CERN to learn a little about the accelerator and the detector.

Our first stop was the Prevessin site, 5km North of the main site. This is where the LHC control room is sited, and this gives a sense of the scale of the operation. The monitoring systems for the accelerators are based in a huge room, the size of an Olympic swimming pool and filled with computers. There is a complex of four different rings, each with their own quirks and purpose, each with its own corner of the room, its own team of experts and its own bank of computers. Building an accelerator is an expensive task, so at CERN the old accelerators are recycled, giving decades of experience from other experiments which are now long gone. The tunnels the LHC uses were built for different purposes, and they delivered fantastic results, as well as several few Nobel prizes. But maintaining the machine isn't easy, and there had been a few teething problems with the accelerator over the past week. So the accelerator physicists were busy, tuning beams, adjusting magnets and doing everything they can to deliver stable beams.

Next we got to see some of the test beam apparatus. In order to test our hardware we need to expose it to high energy beams in a controlled environment. To do this, the LHC delivers some of the beam to the test beam area, a vast warehouse, filled with hardware unique to each experiment. This is serious business- there are huge concrete blocks surrounding each test region to shield physicists (and other equipment) from harmful radiation. We needed to get a better idea of what work was being done, so we walked along the gantry, over the top of the test beams, while the industrial size crane moved above us, shepherding equipment from one area to another. This was our chance to see just how much infrastructure was need to operate one beam and to test one piece of equipment. There was a maze of cabling and tubing coming from every part of the room, and no space was wasted. One end of the room looked just like a warehouse, with calibrated equipment boxed and ready to be shipped, including a vast crate for NASA.


Tingting and Prof Stroynowski discussing the cabling problem.

But still this wasn't quite what we were looking for. We wanted to see something closer to the ATLAS detector. (We can't see the detector itself at the moment as the beams are on, and the detector environment is a dangerous place during periods of data taking.) Instead we went back to the main CERN site to see where the cabling problem was first overcome. To find a way to get all the cables out of the inner detector systems, a life size model of the ATLAS detector was constructed from wood, and the cables attached, one by one, to make sure there was enough space to fit everything in. Seeing these cables finally brought home the true nature of the problem. There were lots of cables of all different kinds- some carried data, some carried power, some added redundancy to the detector, some were used to monitor the detector conditions. When you have that much hardware in one small space there are a lot of things that need cables! And every time you have cables you introduce two new problems. First, the cables need space to get out of the detector, which is why Tingting saw the "gaps" in the liquid argon calorimeter. Second, any particles we hope to see may have to pass through the cables, losing energy as they do. This second effect changes the properties of the particles in ways which may make it very hard, or even impossible, to identify the particles properly, which could essentially make some of the data we take useless.

These problems are difficult to overcome and make the development of the hardware expensive. The shortest answer to Tingting's question may be a simple answer, but the longer answer tells us a great deal about our hardware and how the experiment operates. These discussions are not idle speculation, as physicists spend a great deal of time studying the detector in detail to see how we can get the best from the data, and how we can best understand how the limits of the hardware affect our data. The more we understand the quirks of the various parts of the data, the better our measurements will be, and the more impressive our discoveries will be. In last week's SMU meeting, another grad student, Rozmin, showed us her work on the Z boson, and it was no surprise to see that she too had to deal with these challenges. She needed to properly reconstruct electrons that came from the Z boson, so she had to exclude the regions of the detector with the gaps. Knowing where they were and how they affected the measurements allowed Rozmin to make an informed decision, to get the best from her data and to present a good study of the Z boson.

Taking time away from writing papers, creating simulations and analyzing data, so that we can understand the hardware is a vital part of being a good physicist. Working with data and discovery is exciting, it's why we're here, and it's the reason 2,000 physicists have joined the ATLAS experiment. But without a good understanding of the LHC and the ATLAS detector there is only so much we can extract from the data. This is why we're keen to learn as much as we can about the hardware. It's exciting, it's tangible and it makes the research all the more fascinating.

TrackBack

TrackBack URL for this entry:
http://blog.smu.edu/mt-tb.cgi/11569

About

This page contains a single entry from the blog posted on July 9, 2010 4:51 AM.

The previous post in this blog was The Routine.

The next post in this blog is The many faces of eating at CERN.

Many more can be found on the main index page or by looking through the archives.