April 22nd, 2013

ATLAS

The ATLAS experiment is, along with CMS, one of the two general-purpose particle detectors at the LHC. ATLAS is located within a short walk of the CERN campus, almost diametrically across the LHC ring from CMS. Its dimensions justify the use of “compact” in the name of CMS. ATLAS extends 45 metres along the beam line and has a diameter of 25 metres. It is, however, lighter, tipping the scale at a mere 7000 tonnes. The name “ATLAS” is one of the most awkward acronyms of all time—it stands for A Toroidal LHC ApparatuS. Since “LHC” is itself an acronym, this is a two-level nested acronym. It's a pity “hadron” isn't an acronym in which the “a” stands for “atlas”—then the acronym would be infinitely recursive. Maybe next time….

The side of the above-ground support building is painted with a cross-sectional depiction of the underground detector while the end depicts a cross-section through the detector at the interaction point.

The ATLAS control room has much more a “mission control” vibe than that of CMS. No, the detector is not operated by ghosts: the phantom faces you see are reflections of members of our group in the glass enclosing the control room.

We're off to see the detector!

Like CMS, the detector has been disassembled for maintenance. To the right is the end-cap with its muon chambers. In the middle is the cap which covers the ends of the eight toroidal magnets which provide the large-scale magnetic field within the detector, situated outside the calorimeters and within the muon detectors. A smaller solenoid magnet, not visible here, encloses the inner detector. All of these magnets are superconducting, chilled with liquid helium when in operation.

Looking to the left toward the interaction point (which is buried deep within the detector and not visible from this viewpoint) three of the eight toroidal magnets are visible: they are the white cylinders with orange circumferential stripes. These magnets make a 90° turn as they approach the edge of the detector and continue along its axis away from the camera. The structure at the other end of the detector is symmetrical.

Here is a closer view of the muon system in the end-cap.

To give a sense of scale, note the two workers in white hard hats on a work platform near the centre of the frame. The shaft above extends to the surface. The detector was lowered, piece by piece, down this shaft and reassembled in the cavern.

For full details of the design, construction, and operation of the ATLAS experiment, see the technical paper [PDF] published in 2008 by the ATLAS Collaboration.

For the LHC to function its geometry must precisely conform to the design. Geodetic reference points like this occur at frequent intervals. They mark co-ordinate points transferred underground from precision surveying done on the surface.

Managing the fluids (liquid and gaseous helium and nitrogen) used by the detector requires a great deal of plumbing, a tiny fraction of which is visible here.

When the LHC is operating, every 25 nanoseconds bunches of particles collide at the interaction point, producing an average of 20 collisions, all of which spew out particles detected by ATLAS. A vast, multi-level hierarchy of custom electronics and high-performance computer clusters is required to cope with this firehose of data. Before filtering, it amounts to one petabyte (10¹⁵ bytes) per second. Even after discarding “uninteresting” events and compression, the detector generates 100 megabytes of data per second, all of which must be stored for off-line analysis. I have no idea which part, if any, this modest server room plays in the process; it's the only one we saw on our tour.

by John Walker May 12th, 2013

This document is in the public domain.