The Particle Accelerator That Could Tell When the Train to Paris Left
It is two in the morning, and one of the physicists running Europe's biggest particle accelerator is sitting in a parked car in the Jura mountains, watching the sky for airplanes. Steve Myers has a theory: something about planes landing at Geneva airport is jostling the energy of the particle beams in his 27-kilometer machine. But it is too dark to see much, and the last flights landed hours ago. His verdict, as he later wrote it: "Experiment inconclusive!" The real culprit ran on rails, and it was bound for Paris.
By the Curiosities Desk · · 5 min read · Filed under Physics · Today’s round
What happened: The Particle Accelerator That Could Tell When the Train to Paris Left
The Large Electron–Positron Collider, or LEP, switched on at CERN in 1989. It filled a 26.7-kilometer tunnel beneath the French–Swiss border near Geneva, the same tunnel that holds the Large Hadron Collider today. LEP sent electrons and positrons (the antimatter twins of electrons) racing around the ring in opposite directions and smashed them together to make Z bosons, heavy particles that carry the weak nuclear force.
One of LEP’s main jobs was to weigh the Z as precisely as possible. That meant knowing the energy of the beams with extraordinary accuracy, because the collision energy where Z production peaks reveals the particle’s mass. The physicists had a clever tool for this, called resonant depolarization. Over time, the spinning electrons in the beam line up on their own. Tickle them with a magnetic field oscillating at just the right frequency and that alignment collapses, and the frequency that does it tells you the beam energy to within a fraction of a million electron volts (MeV), out of roughly 45,600 MeV per beam.
The trouble was that the energy refused to sit still. The first calibrations, in 1991, showed fluctuations nobody could explain. A ground-motion expert from SLAC, the Stanford accelerator lab, suggested the Moon. The solid Earth has tides, too: its crust rises and falls slightly as the Moon and Sun pull on it. In November 1992, during a new moon, the team ran a dedicated experiment and watched the beam energy swell and ebb with the tide. The ring’s circumference was changing by about a millimeter, out of 27 kilometers. Later, the team found slower, seasonal flexing too. Rain and meltwater from the mountains filled Lake Geneva and, as Myers put it, weighed down one side of the ring.
Then came a second mystery. Myers recalls that in 1993 the energy signal showed random jitters during the day but went quiet for a few hours in the middle of the night. Everyone had a pet theory. His was aircraft, which led to the 2 a.m. car-park stakeout. A few weeks later, a chance conversation with a railway engineer cracked it: the TGV, France’s high-speed train.
Electric trains on the line running from Geneva toward Bellegarde, on the route to Paris, drew direct current, which is supposed to return through the rails. Some of it leaked into the ground and found an easier path home: LEP’s long metal vacuum chamber, the pipe the beams travel through. According to Myers, the current flowed along the accelerator and returned through the Versoix river to Geneva’s Cornavin station. A later CERN talk put the stray current on the vacuum chamber at about 1 ampere. The quiet hours were simply the hours when no trains ran. And the very first energy measurements had looked clean, Myers wrote, because TGV workers had been on strike.
When the team installed magnetic-field probes inside some of the bending magnets in 1995, they caught the effect directly. Through each run, the magnets’ field crept upward, and with it the beam energy, by around 5 MeV.
Why it was strange: The Particle Accelerator That Could Tell When the Train to Paris Left
LEP was built to probe particles far smaller than an atom, dozens of meters underground and sealed off from the world above. Yet it turned out to be sensitive to the Moon, the level of a lake, and the French railway timetable. A 1999 article in CERN Courier joked that LEP’s energy calibration had become “an expensive way of monitoring the passage of French TGV high-speed trains.”
It is also a lovely case of a strike doing science a favor. The trains were absent exactly when the physicists first looked, so the noise only appeared later, as if it had started on its own.
Another lab signal that turned out to have a very ordinary source: The Kitchen Microwave That Impersonated a Signal From Deep Space.
What scientists learned: The Particle Accelerator That Could Tell When the Train to Paris Left
The trains taught a lesson about magnets. The leaked current gradually magnetized the iron in LEP’s bending magnets, nudging their field up by about one part in 10,000 over a run. Since those magnets set the beam’s path, the energy rose with them.
The tides taught a lesson about geometry. In a ring this size, a shift of one millimeter forces the beam onto a slightly different route through the focusing magnets, and that alone changed the energy by roughly 10 MeV.
Once each effect was understood, it could be modeled and corrected, and the corrections shifted the Z mass by a few MeV. The final combined value from LEP’s four experiments and SLAC was 91.1875 billion electron volts, give or take 0.0021, a precision of about two parts in 100,000. It remains one of the best-measured numbers in particle physics, and it only came out right because someone accounted for the Moon and the trains.
Another case where a tiny hidden glitch steered a big physics result: The Loose Cable That Made Neutrinos Outrun Light for Five Months.
How it reaches us today: The Particle Accelerator That Could Tell When the Train to Paris Left
The LHC, which now occupies LEP’s tunnel, adjusts its radio-frequency system to compensate for the same Earth tides. And the planning for CERN’s proposed next machine, the Future Circular Collider, leans directly on LEP’s experience. Its energy-calibration working group recommends measuring the beam energy every few minutes during data taking and building a detailed model of everything that can drift, from magnet fields to the size of the ring. Its target uncertainty for the Z mass is about 100 thousand electron volts, roughly 20 times tighter than LEP’s.
The bigger lesson reaches well beyond physics: the more precisely you measure something, the more of the world leaks into the measurement. A 2017 CERN talk on the topic noted that the Geneva–Bellegarde line had since been converted to alternating current, retiring one source of stray currents.
More at CERN where the outside world got into the machine: The Baguette Blamed for Overheating the Large Hadron Collider.
Fun fact: The Particle Accelerator That Could Tell When the Train to Paris Left
Trains weren’t LEP’s only intruders. In June 1996, Myers found an obstruction in the beam pipe that looked like the bottom of a beer bottle. It was: a Heineken bottle, and a second one turned up too. The culprit was never identified.
Sources: The Particle Accelerator That Could Tell When the Train to Paris Left
- Steve Myers, “The greatest lepton collider”, CERN Courier, September 11, 2019.
- The LEP Energy Working Group (R. Assmann et al.), “Calibration of centre-of-mass energies at LEP1 for precise measurements of Z properties”, The European Physical Journal C 6 (1999).
- ALEPH, DELPHI, L3, OPAL and SLD Collaborations, “Precision electroweak measurements on the Z resonance”, Physics Reports 427, 257–454 (2006).
- J. Wenninger, “Energy Calibration and Stray Fields at LEP: Spins, Tides & Vagabond Currents”, CERN Mini-Workshop on Stray Fields, October 2017.
- FCC-ee Energy and Polarization Working Group (A. Blondel, P. Janot, J. Wenninger, eds.), “Polarization and Centre-of-mass Energy Calibration at FCC-ee”, arXiv:1909.12245 (2019).