Recent Posts
Musings on the Large Hadron Collider
11/12/2009I was consistently vexed by many folks on Hexus referring to the recent proton-beam collisions achieved in the LHC as ‘laser’ beams.
This was my response,
I hope folks understand the complete achievement in #epicfail by referring to a beam of sub-atomic particles as a laser. A laser by definition is light amplification by the stimulated emission of radiation. While it is true you can look at visible light as either the easily understood wave model, there is also the particle model (photons etc). But that does not change the fact that visible light is electromagnetic radiation – an electromagnetic wave that has both a sinusoidal E-field (electric field) and B-field (magnetic flux density, as we engineers call it; physicists just call it the magnetic field) oscillating in phase at right angles.
Now, going back to the laser, this is simply a device that emits light, albeit highly amplified, at a singular wavelength, which is why they are characterised by various colors – with the advent of blue lasers allowing more data to be etched onto the surface of BluRay DVDs for example. While there are very powerful military grade lasers about, I do not suspect any of them can transmit the same quantifiable amount of energy as described by Gordon.
I just guess it is ‘easier’ to conceptualise it as a laser, as that is probably the only beam most people can visualise in the first place.
I found some fantastic images online, mainly thanks to Fraz’s continued LHC discussions on Hexus. He explains,
This was a very low-energy collision, but a collision nonetheless! If you compare against pictures I posted a while back in this thread, the above picture looks a lot more normal – i.e. the beam isn’t hitting the side of the detector, but stuff is colliding at the centre instead. The outer part of the silicon tracker is now also switched on, which is why there are some properly reconstructed tracks (yellow lines) now seen that weren’t in previous images. We couldn’t include the tracker before from fear of damaging it with a poorly controlled beam, which is the reason why the inner tracker is still currently off.
Progress Update on Recent LHC Activities
Fraz says,
Things have moved on a bit, but not much – to me, it seems that the main thing going on is a lot of champagne drinking…
We’ve been colliding very low-luminosity beams (low luminosity means not many protons in the beam) at the injection energy, which is 450 GeV per beam. So basically we’ve just been storing beams in the LHC that have been accelerated in an earlier synchrotron in the chain of LHC pre-accelerators, and then crossing the beams every now and again.
We’ve also used the LHC to accelerate the beams a small amount (up to 540 GeV), but have yet to cross any of these slightly higher-energy beams.
Some people have been trying to analyse the very early data to see how the detectors are doing. Amusingly on one mailing list, some guy is claiming he has “rediscovered the kaon” in one of the events (it was originally discovered in the 1940s), so it looks like the detector I work on is performing well enough to do this, at least!
I think the priority right now is trying to improve the beam lifetime. The LHC is a storage ring as well as an accelerator, and should eventually be able to store 7 TeV beams for ~10 hours. Currently I think we’re only managing to store the beam for about 40 minutes, and that’s only at 450 GeV per beam. Once beam lifetime is improved, we’ll move on to trying to take the “Highest Energy Particle Accelerator” crown from Fermilab. Fermilab has 1 TeV beams, and thus 2 TeV collisions. We’re aiming to exceed 1TeV beams sometime in December.
Data Storage from an Event
An automated magnetic tape vault at CERN computer center, seen on September 15th, 2008. The tapes are used to store the complete LHC data set, from which a fraction of the data is copied to overlying disk caches for fast and widespread access. The handling of the magnetic tape cartridges is now fully automated, as they are racked in vaults where they are moved between the storage shelves and the tape drives by robotic arms.(Claudia Marcelloni; Maximilien Brice, © CERN)
Fraz explains,
Yeah, that system is called CASTOR. It’s good and bad. In certain circumstances, you can certainly be waiting for quite a while for your data to be staged on the hard-disk front-end. I guess that’s pretty unavoidable though.
Raw data alone – i.e. before it gets turned into more useful physics analysis objects – the CMS experiment that I’m working on will be producing about 10 peta-bytes per year. Once you factor in converting the raw data into more useful formats, add on some Monte-Carlo simulation data, and then some data duplication to 2 or 3 other large centres (such as Fermilab), it’ll be pushing 100 peta-bytes per year.
The ATLAS experiment will be producing about the same amount. I’m not sure about the two other smaller experiments, LHCb and ALICE, but let’s call them about 100 peta-bytes together.
So… ball-park, the LHC experiments altogether will be producing about 300 peta-bytes of data per year, and we’ll be running for about ~10 years before we upgrade things. So, for the lifetime of all current LHC-related CERN experiments, I guess the total dataset will be in the region of 3 exa-bytes.
“Any backup?”
Yes, data duplication with other major data centres, such as the one at Fermilab. Basically CERN forms the “Tier 0″ data centre, where one copy of the entire dataset will be stored. There are then a number of “Tier 1″ data centres, such as at Fermilab in the US and the Rutherford Appleton Laboratory here in the UK. Between all these Tier 1 centres, the data will be duplicated, but not all of it at any single Tier 1 centre. Then there are Tier 2 centres at various universities, which again between them duplicate all the data at each Tier 1. So, it should be fairly safe.
LHC: Storing Beams!
“I didn’t realise it could be used to store the beams as well, but thinking about it, it’s fairly logical that it would be able to do that (accelerates it, then just keeps the energy cycling, like a giant centrifuge?)”
Yeah, basically that’s right. Right now, the beams aren’t living very long because we don’t have a tight enough control over them. But when we do have good control of the beams, the reason they will only lasts for ~10 hours is because we’ll be crossing them at four locations on the ring every 25 nanoseconds. This slowly degrades the number of protons in the beam, until eventually we discard it and create a new one.
When we’re running at full steam, the beams will consist of about ~3000 tight bunches of protons spaced 25 nanoseconds apart in time, with each bunch containing ~200 billion protons. Every time we collide the beam, roughly 20 proton-proton collisions will occur… so naturally the luminosity of the beam degrades over time because some protons are lost.
So… you see that there are a lot of protons in the beam, each of which has a lot of energy for something so small. To give you an idea, when things are at full power in a couple of years or so, each circulating beam would be capable of instantly liquefying 500 kilos of copper. So, building a beam dump that can absorb the beam if we need to get rid of it quickly is quite difficult… I think each one weighs in at 1000 tonnes and they have to be water cooled to disperse the heat quickly enough.
urrently we are running a low luminosity beam, but at full luminosity the stored energy in the beam is huge. Silly factoids are as follows:
At full luminosity & proton-energy, the beam will store 362 mega-joules of energy, which is:
1) The same energy as the HMS Invincible aircraft carrier moving at 12 knots.
2) The same energy that a Subaru Impreza moving at ~3000 mph would have.
3) Enough energy to melt half a ton of copper.
4) The equivalent of 77.4 kg of TNT. [Correction: it's 87 kg of TNT]Then bear in mind that the LHC could deliver all this energy into an area much smaller than a single square centimeter in less than 90 microseconds.
LHC Photos
The silicon strip tracker of the Compact Muon Solenoid (CMS) nears completion. Shown here are three concentric cylinders, each comprised of many silicon strip detetectors (the bronze-coloured rectangular devices, similar to the CCDs used in digital cameras). These surround the region where the protons collide. (© CERN)
View of the CMS Detector before closure, on August 17th, 2008. (Maximilien Brice; Michael Hoch; Joseph Gobin, © CERN)
Installation of the mini frame of ALICE on 15 May 2009. (Maximilien Brice; Mona Schweizer, © CERN)
Installing the ATLAS calorimeter in November of 2005. The eight torodial magnets can be seen on the huge ATLAS detector with the calorimeter before it is moved into the middle of the detector. This calorimeter will measure the energies of particles produced when protons collide in the centre of the detector. (Maximilien Brice, © CERN)
You can view many more stunning images of the LHC here.
There are 2 comments in this article: