Sophisticated techniques like page-flip buffering enable analog-to-digital converters to offload data and re-arm at frequencies as high as 50 Mhz.
Last summer’s observation of a particle that just might be the elusive Higgs boson triggered a frenzy of excitement worldwide. It took another six months, however, for scientists at CERN’S Large Hadron Collider (LHC) to confirm the discovery. The time lag serves to underscore the complexity of the task-not just measuring the results of the collision but the process of triggering the collisions in the first place. In order for the experiments to succeed, a complex chain of tasks has to take place with perfect synchronization-from launching the particles into the accelerator to digitizing the data from the location sensors used to pinpoint the particles so that they can be brought up to energies as high as 7 TeV.
CERN, the European Laboratory for Particle Physics, is far more than just the LHC. The facility includes 10 different accelerators designed to bring different particles to specific energy levels. For the Higgs boson experiments, a network of four different accelerators successively boosted the energy of the proton beam until it entered the LHC, where half circulated in the clockwise pipe and half were split off to the counterclockwise pipe, then brought together in the target chamber to collide.
Although the term proton beam is bandied about, the protons actually circulate in spatially localized bunches of particles with picosecond widths. The energies involved are boggling. Protons start in the linear accelerator Linac 2, which raises their energy to 50 MeV (see figure 1). Next, the packet stream circulates through the Proton Synchrotron Booster, to accelerate to 1.4 GeV. The particles are injected into the Proton Synchotron, which boosts the energy to 25 GeV, then the Super Proton Synchotron, where they are accelerated to 450 GeV. The last step is the LHC, which takes them to energies as high as 7 TeV and speeds of 0.999999991c
; at that point, making a loop of the 27-km-circumference LHC takes just 90 ms.
Figure 1: Accelerating protons to tera-electron-volt energies requires
a network of accelerators that successively boost
particle speed ever higher.
Accelerating particles up to tera-electron-volt levels is a complex and precisely choreographed dance. Boosters throughout each accelerator apply carefully timed bursts of RF energy to the particle bunches as they pass by. With the exception of the Linac 2, the CERN accelerators are rings that allow packets to circulate multiple times until they reach the desired energy; in the case of the LHC, for example, this takes around 20 minutes. The transit of a single accelerator ring can range from microseconds to nanoseconds in time. At exactly the right moment, apertures between accelerators open to transfer packets from one accelerator-and energy regime-to another.
It’s a tricky process. The collisions may be discrete events, but data acquisition periods can last as long as 24 hours, during which time the packet streams continue to circulate. The problem is that over time, the packets tend to spread out. The RF pulses can be used to compact them, but this involves a complex process of slowing down the leading particles and speeding up the lagging particles, a task complicated by the fact that the entire bunch is spread out across just picoseconds. It’s an exacting process that requires precise, accurate position data.