Errant baguette chunk shuts down collider

[snopes]

Experts are trying to determine how a piece of bread made its way into the Hadron collider in Switzerland, shutting it down for a couple of days.

http://www.upi.com/Odd_News/2009/11/...4651257482954/

[Nick Theodorakis]

I thought for sure this was going to be an article from The Onion.

Perhaps the baguette was sent from the future.

Nick

[Magpye]

I have no theories, but I find this completely hilarious.

[Jay Tea]

It is amusing - I wonder what the British version of this would be - 'Egg and Cress sandwich crust shuts down Diamond Synchrotron'

[BlueStar]

They're playing Russian Baguette with the universe!

[jimmy101_again]

Originally Posted by Jay Tea

It is amusing - I wonder what the British version of this would be - 'Egg and Cress sandwich crust shuts down Diamond Synchrotron'

US West coast version:
'SLAC shutdown by errant In-n-Out Burger fries'

US Midwest version:
"Fermilab accelerator shutdown by errant White Castle slider"

[Eddylizard]

The best guess is that it was dropped by a bird, either that or it was thrown out of a passing airplane,"

Okay how do you throw a crumb out of an aircraft, let alone get it to land somewhere in an underground facility? How fragile was the

electrical unit that should have been powering the cryogenic cooling unit.

I'm pretty sure the electrical unit that supplies power to my house could sustain multiple and repeated attacks by full sized intact baguettes.

Just asking.

[snopes]

A Cern spokeswoman, Christine Sutton, said scientists had headed above ground to investigate when they made their discovery.

"The problem related to the high voltage supply," Sutton said. "We get mains voltage from the grid, and there was an interruption in the power supply, just like you might have a power cut at home. The person who went to investigate discovered bread and a bird eating the bread."

Sutton said the bird and its bread were discovered at a compensating capacitor – one of the points where the mains electricity supply enters the collider from above ground.

The incident cut power to one of the collider's cooling plants, causing temperatures to rise by more than 3C in part of the tunnel.

http://www.guardian.co.uk/science/20...bang-goes-phut

[Mr. Billion]

The LHC is surprisingly fragile.

[Tarquin Farquart]

It's a baguette from the future, designed to prevent any experiments taking place.

[E. Q. Taft]

I am reminded of Harry Harrison's Star Smashers of the Galaxy Rangers, in which a mischievous boy genius arranges for a piece of cheddar cheese to be bombarded with charged particles as a prank -- but inadvertently invents the "Cheddite Projector," which enables faster-than-light travel....

Hey, it could happen!

[Mad Jay]

Haven't these people heard of the term "redundant systems"?

[jimmy101_again]

Originally Posted by Mad Jay

Haven't these people heard of the term "redundant systems"?

I'm sure they have.

I'm sure they have also heard the term "budget". Want to guess which of those two terms the builders hear more often?

Redundancy is most often associated with keeping people alive. Manned spaceships are highly redundant system. Unmanned systems much less so.

The collider needs redundancy based on how dangerous it is if there is a catastrophic failure. I believe the magnet in the collider are supper cooled. Loss of power is a very dangerous occurence since the supercooled magnets will experience a "temperature excursion" if they get too warm. The pressure explosion resulting from a magnet "excursion" may well kill people that are in the vicinity.

But I doubt anyone is anywhere near the magnets when the collider is powered up. Quenching of the magnets would be the least of a persons worries in the immediate vicinity of the device.


... personally, I suspect the folks at Fermi lab. After all, you can get a loaf of bread through security at an airport and at the collider.

[Alchemy]

Originally Posted by jimmy101_again

Redundancy is most often associated with keeping people alive. Manned spaceships are highly redundant system. Unmanned systems much less so.

Additionally, the redundancy tends to be toward ensuring proper detection of situations where there is danger of harm to a person or risk of catastrophic damage to the equipment and shutting everything down in a safe manner.

I believe the magnet in the collider are supper cooled. Loss of power is a very dangerous occurence since the supercooled magnets will experience a "temperature excursion" if they get too warm. The pressure explosion resulting from a magnet "excursion" may well kill people that are in the vicinity.

No superconducting magnet would ever be (deliberately) installed such that a quench would allow its cryogenic reservoir to reach burst pressure. (Not to say it's *impossible* for this to happen, it's just an obvious danger and designed/tested against.) Further, magnets this large are designed with a large-diameter quench vent to deliver its cryogenic gases out of the building. (Although if this is blocked, these gases will be vented in whatever room the superconducting magnet is in.)

But I doubt anyone is anywhere near the magnets when the collider is powered up.

Primarily because this would place a person inside the radiation shield.

[DemonWolf]

Originally Posted by jimmy101_again

I believe the magnet in the collider are supper cooled.

Maybe that's how it got in there. it was part of the supper coolant?

[Mad Jay]

Originally Posted by jimmy101_again

I'm sure they have.

I'm sure they have also heard the term "budget". Want to guess which of those two terms the builders hear more often?

Redundancy is most often associated with keeping people alive. Manned spaceships are highly redundant system. Unmanned systems much less so.

Umm no. Most web sites run on redundant servers. Chances are this very website is running on redundant servers. Most server farms have multiple redundant power supply systems and cooling systems. If you go to any web hosting providers that promises 24x7 uptime, they probably have redundancy in everything. Redundancy is used everywhere. Nobody cares about it unless there's risk to life and limb, but they are everywhere. Not having redundant systems is just bad engineering

As far as budgeting goes, usually the way the cost for these things is computed, is they compute the cost for downtime vs the cost of having a redundant component. Let's say, you have a cooling system that historically fails 80% of the time, and it costs 10K a year to run it. This cooling system cools a part of a system, that costs let's say 100K a year to run. That means if the cooling system fails 20% of the time causing the larger system to halt 20% of the time, which means that the failure in the cooling system costs them 20K a year. It makes sense to have 2 cooling systems and guarantee near 100% uptime

THe point I'm making is that the LHC grinding to a halt because a bird sat on a wire eating a piece of bread which caused a cooling system to fail is bad design. It's not goobleydook from the future. I am pretty sure it costs them a hell of a lot of money trying to track the problem than to actually fix it.

[Alchemy]

Originally Posted by Mad Jay

Umm no. Most web sites run on redundant servers.

The LHC is not a web site.

Particle accelerators do not utilize redundant rings, or redundant supercooled superconducting magnets. These systems are not designed for 100% up-time. They are designed for running short-term (hours, minutes, picoseconds) scientific experiments. The most-worked particle accelerators are open for experiments about 20 days a month. The RHIC at Brookhaven, a collider very similar to the LHC, has an up-time of around 50%.

Redundancy is utilized in the devices which protects lives and equipment. Redundant instrumentation and uninterruptable power supplies are dedicated towards quench protection and other critical events. Redundancy is not used to ensure perpetual operation. High-energy experimental physics facilities are not open for experiments 24/7/365.

[Mad Jay]

Originally Posted by Alchemy

The LHC is not a web site.

Particle accelerators do not utilize redundant rings, or redundant supercooled superconducting magnets. These systems are not designed for 100% up-time. They are designed for running short-term (hours, minutes, picoseconds) scientific experiments. The most-worked particle accelerators are open for experiments about 20 days a month. The RHIC at Brookhaven, a collider very similar to the LHC, has an up-time of around 50%.

Redundancy is utilized in the devices which protects lives and equipment. Redundant instrumentation and uninterruptable power supplies are dedicated towards quench protection and other critical events. Redundancy is not used to ensure perpetual operation. High-energy experimental physics facilities are not open for experiments 24/7/365.

There is no need for something to be operating 24x7 for that thing to require redundant systems. Redundancy is one way you can achieve 24x7 uptime, but that doesn't mean you use redundancy only in systems that require 24x7 uptime.

It's all about cost. The basic principle is that if the cost for downtime is high, then you add redundant components, as long as the components are cheap enough (No, you can't have 2 LHCs! )

[jimmy101_again]

Alchemy is probably correct, the magnets would be designed to quench fairly safe and an explosion is probably unlikely. In addition, the refrig system probably isn't even needed to keep the magnets superconducting cold.

On the other hand, it's not like these are production magnets. The safety features would have been tested but a device really hasn't been properly tested until it's been in service for several years. For example, the transformer that the baguette and bird was interfering with.

Still, dumping many hundreds(?) of gallons of cryogenic fluid (helium?) into an enclosed space in a short period of time is a problem in itself, even with proper venting to the outside ... exactly how deep is the accelerator buried??

[Alchemy]

Originally Posted by jimmy101_again

On the other hand, it's not like these are production magnets. The safety features would have been tested but a device really hasn't been properly tested until it's been in service for several years.

Indeed. And the incident that put the LHC out of business last year was a quench event not unlike what you mentioned - the root of the problem wasn't that the magnet quenched, but that an element arced, physically compromising the cryogenic vessel and allowing helium to enter the insulating vacuum layer. The fault triggered an automatic quench. The resulting pressure increase was safely handled by the cryogenic pressure relief valves, but well beyond the design of the vacuum pressure relief valves. The overpressure in the vacuum system caused tremendous mechanical damage.

Still, dumping many hundreds(?) of gallons of cryogenic fluid (helium?) into an enclosed space in a short period of time is a problem in itself, even with proper venting to the outside ... exactly how deep is the accelerator buried??

The previous incident released 2 metric tons of helium in the initial quench and another 4 tons by slow leak, out of a total of 15 tons in that section. I think the entire LHC requires a total of 100-120 tons. You're absolutely right that the access tunnel (off limits during operation) was rapidly rendered incompatible with human life and remained so for days.

PDF: https://edms.cern.ch/file/973073/1/R...at_LHC__2_.pdf