The Large Hadron Collider, small step or giant leap?

Monday, 02 November 2009

Earth, Nov. 2009 I’m off to Oz tomorrow, and as they’re likely to fire it up whilst I’m away, I thought it best to fire out another LHC- post.

Last time I gave a run-through of the concepts and science involved with that there hadron smasher, but now I’m going to talk about some of the engineering involved

(which is much more in my comfort zone) and the issues encountered which have postponed doomsday until a few weeks’ time. I’ve purloined some material from work to illustrate. Special thanks to Dr Bruce Kennedy -who works on CMS here at RAL- for the material.

Here’s a video to get us started:

Quick recap: What’re we doing at the LHC again?

Well, historically we have been seeing the form and nature of atoms using instruments such as electron microscopes. If we wish to see inside atoms and their component parts, we need to resolve much higher energies than that kind of equipment can provide. For example, here is a proton:

If we wanted to resolve the structure of this proton we need to use a beam with energy of at least 1 GeV (eV=electron volt: a unit of energy which is also a unit of mass in particle physics, confusingly). Unfortunately this is enough energy to smash the proton to pieces. So the best we can do is commence with the smashing, and look at those pieces and determine the make-up of the proton from that.

How do we do that? Well first off, we need to create some protons. At the very start of the LHC process there is an ion source which produces H+ ions (which are your standard hydrogen atoms with the electron stripped away, leaving you with a proton). Because protons have an electrical charge, they can be accelerated and directed by applied electric fields.

The protons are injected into the Linac (LINear ACcelerator), where they are accelerated using 129 pulsed quadrupole magnets. These magnets bunch the protons together, focussing and accelerating them in a straight line towards the first of several synchrotrons. A synchrotron is a circular arrangement of magnets and RF cavities which accelerates particles round and around until they reach a certain energy, at which point they are generally kicked out for some other purpose, or they can, like the electron-based machine at DIAMOND (next door to me at RAL), just keep the particles spinning over and over to produce light energy for their experiments. Alternatively we can go the LHC route and just smash beams together and see what happens.

Where were we? Ah yes, we’ve just been fired out of the Linac. Now the first synchrotron the protons encounter after they leave the Linac is the Proton Synchrotron Booster, into which they are injected with energies of 50MeV. Here they are accelerated further until they reach energies of 1.4GeV, at which point they are fed into the Proton Synchrotron(PS) which accelerates the protons further still until they reach 25GeV. Can you guess what happens next? Yes, they get fed into yet another synchrotron called, imaginatively enough, the Super Proton Synchrotron (SPS).

Now the SPS on it’s own is pretty massive, coming in at 7km around (compared to the ISIS synchrotron which is about 163m in circumference), and it is the base for several existing science programs like COMPASS and CNGS. The SPS spins our long-suffering protons up to 450 GeV, before sending them on their merry way into the LHC tunnels.

This is the fun part. Every other ‘bunch’ of protons is sent in the opposite direction to the last, around the parallel rings of the LHC. As I mentioned last time, they are sped up to within a tiny fraction of the relativistic speed of light before being steered into each other’s path. The energy of each beam is set ultimately to be 7TeV, meaning a cumulative energy of 14 TeV when they collide (although they are going to start at a much more modest 1TeV and work up from there). The collisions are created amidst large detector arrays, like ATLAS:

Atlas

I will attempt to embed a flash animation now to illustrate the whole thing. No-one move or breathe (right clicking and zooming might be useful):

So what went wrong when they tried to switch it on last October?

Well the the ring of the LHC itself consists mainly of a chain of magnets –which in turn are broken into two basic types: dipoles for steering and quadrupoles for accelerating. Together the LHC’s magnets store considerably more energy than the beam itself; about 10,000 megajoules compared to 362 megajoules for the beams. Most of this energy is contained in 1232 superconducting dipole magnets. ‘Superconducting’ means they have been cooled down to the point where electrical resistance in the magnet coils is effectively zero, allowing for the currents needed to produce the massive magnetic fields we’re looking for. And by cold we mean 1.9 kelvin (-271C; -456F).

So they were testing the powering of some dipoles in sector 3-4 last September, and they encountered a bit of a problem. Well actually it was a lot of a problem. The system suffered a Quench, which is where a magnet is inadvertently heated beyond a critical point, changing it from superconducting to just, well, conducting. This change releases the stored energy of the magnet and that of the neighbouring magnets too, which can have rather dramatic effects.

In this case the quench was caused by an electrical short, meaning that a power transformer on one of the surface points of the LHC switched off the main compressors of the cryogenics for the two sectors of the machine, compromising the integrity of the cryogenic system (this sector was one of the last to be done – managers take note: keep an eye on your subbies towards the end of a project when the pressure’s on to finish ASAP). The consequences were pretty severe:

The fault caused a circuit to overload, causing an arc which knackered a helium vessel. The helium then leaked into the vacuum jacket surrounding the chain of magnets, causing a massive overpressure. Ultimately, 24 dipoles and 5 quadrupoles were ripped out of their fixtures and thrown several feet down beam. the floor was smashed and 6 tonnes of helium was vented into the tunnels. It was what is known in the trade as 'borked'. But don’t take my word for it:

That wrote off the whole facility for a year.

So now, after a massive amount of repair work and several extra tiers of system integrity failsafe installation; they’re ready to go again. In fact they’re already running the beam up in preparation:

They might not know it, but this will be this generation’s moon-landing. It might seem arcane, unknowable and downright confusing, but this will be -in all likelihood- an epoch-making event in human understanding. So whilst everything else might well be grim, be glad you were there when another couple of pieces of the jigsaw were put in place.

Author: Mr Salted Slug

Trackback(0)

TrackBack URI for this entry

Comments (2) add comment

Show/hide comments

wrinkled weasel said:

	... Having a Quench sounds bad. To a non-geek this all sounds bad. I like Switzerland and Swissies, and do not like the idea of it being swallowed up by a machine. Can I go to Leysin this year in safety?
	report abuse vote down vote up

March 18, 2010 | url
Votes: +0

VioletPETTY said:

	... When you're in a not good position and have got no money to get out from that point, you would have to receive the loans. Just because it should aid you emphatically. I take financial loan every single year and feel myself good just because of this.
	report abuse vote down vote up