The daring success of the world's largest atom smasher on its opening day was more surprising to many scientists than the troubles it subsequently developed.
A problem with a magnet connection will delay the start of experiments for half a year, partly because the $3.8 billion accelerator is so complicated to repair. Physicists — some of whom waited two decades to use the new equipment — will now have to wait three more weeks for the damaged section to be warmed up to room temperature.
They can then get inside to see what went wrong.
Yet such glitches are not uncommon. Michael Harrison, who worked on Fermilab's Tevatron collider and designed and built the United States' other superconducting collider at Brookhaven on Long Island, said both machines had similar startup problems.
"You find all kinds of little issues as you start to turn it on and make it work," Harrison said in a telephone interview from Brookhaven.
Rival operations in the United States watched with admiration as CERN flawlessly launched its new Large Hadron Collider before the world's news media, the first time such a startup has been public anywhere.
"A huge cheer went up from here," said Judy Jackson, spokeswoman for the Fermi National Accelerator Laboratory outside Chicago, whose Tevatron is losing its title of world's highest-energy collider with the startup of CERN's.
Scientists watching the Sept. 10 event live at Fermilab "were all holding our breath," Jackson said. "People here had been through this and said, 'Oh man, I hope it works.'"
The meltdown of a connection between superconducting magnets nine days later at CERN was more expected. But it knocked out use of the machine for two months, costing the scientists valuable time to get the machine ready for use on experiments next spring after the usual four-month winter shutdown.
"It's what happens when you start up a big, superconducting machine," Jackson told The Associated Press. "Our impression is that what happened last Friday at CERN caught them more by surprise than it did us."
Besides the two colliders in the United States, there was a third in Hamburg, Germany, until last year. The United States was building what would have been the largest — the Superconducting Super Collider — but in 1993 Congress stopped construction in Texas after costs soared and questions were raised about the value of the science it could produce.
The machines have to operate at extremely cold temperatures to take advantage of superconductivity — the ability of some metals to conduct electricity without any resistance near absolute zero degrees. That allows for much greater efficiency in operating electromagnets that guide the beams of protons at near the speed of light until the particles collide with each other.
CERN specialists have already figured out that a connector between electromagnets failed and heated up, causing a magnet "quench," or shutdown. It apparently melted a hole in the tube, causing a leak that spilled about a ton of liquid helium.
The collider has well-tested systems to protect against quenches in all 9,600 magnets, and they worked, said CERN spokesman James Gillies. The problem came in the protection for quenches between magnets, he added.
He said the damaged magnets will have to be repaired or replaced.
The use of superconducting components has led to major advances in particle physics because it allows much higher energies to be used in colliding protons and other particles.
The high-energy collisions, which are recorded in cathedral-sized detectors at intervals on the collider, enable physicists to understand better how the smallest bits of matter — and everything and everyone — are made. They also hope it will take them even closer to the "Big Bang," which many theorize was the massive explosion that formed the universe.
By colliding protons from the nucleus of hydrogen atoms at high energy, the CERN machine is designed to recreate, on a minuscule scale, a view of what matter looked like in the rapid cooling one-trillionth of a second after the explosion.
Physicists have used smaller, room-temperature colliders for decades to study the atom. They once thought protons and neutrons were the smallest components of the atom's nucleus, but the colliders showed that they are made of quarks and gluons and that there are other forces and particles. And they still have other questions about antimatter, dark matter and particle mass they hope they can answer with CERN's new collider.
But making colliders superconducting is a major challenge. In the case of the LHC, it means making the 17-mile accelerator ring colder than outer space.
And the warm-up, cool-down time needed to make repairs — the "thermal cycle" — is longer than for Fermilab, where the simpler design of the magnets only requires two or three days. Brookhaven needs a month, Harrison said.
Brookhaven also had problems with connections between magnets that caused a quench.
"The thing about the LHC was it has not just caused a quench, but there are systems that are supposed to prevent it from melting and dumping helium," Harrison said. "So it was obviously something else that went on as well."
CERN has been receiving words of encouragement from rivals, who stressed that startup problems are common.
Jackson said that despite the problems, there have been great strides in the making of magnets. She said Fermilab had lots of problems with exploding magnets when it was commissioning its first, room-temperature accelerator four decades ago, leading the scientist in charge to paraphrase Alfred Tennyson's 1854 poem "The Charge of the Light Brigade": "Magnets to the left of us, magnets to the right of us, magnets in front of us volley'd and thunder'd."
"They were blowing up all the time," said Jackson.