Beam Xover questions

[jmayes]

Most Magnets have 2 tubes for beams to go in both directions simultaneously, My questions is about how the cross over points work and why do I not see 2 tubes going through the experiments too, there must be some bypass system but machine experts say beam travels through the experiments all the time? Is there a good graphic that shows in detail the beam piping system. I work better with schematics then discussion

Great site!
Glad to have found it,
Thankx,
J

[RonnyV]

Here some not too well structured thoughts ...

As you can see in the following link: http://hcc.web.cern.ch/hcc/beam/by_beam.php
the "crossover" points are located at the experiment points (Atlas, Cms, Alice, LHCb).

They are not actually showing as an "X" layout you can see crossover at one point, but in the experiments (imagine the detector size, so we are talking about meter scale here), the beams run through one pipe in opposite directions. Hey, they are running with nearly the speed of light in practically perfect vacuum, so it's not too difficult for each bunch of protons to do this.

Remember the LHC's predecessor, the LEP (Large Electron Positron collider) that was housed in the same tunnel, was running the two beams in the same (oval shaped) pipe. They could do this because the beams of electrons and positrons have different electric charges, so a magnetic field will force the particles in opposite directions. The size of the particle bunch is order of magnitude mm, the beam pipe diameter cm. Now the magnitude order for the area is squared ! So there is lots of room for two beams in one pipe without "hitting" each other (or the wall of the tube, as the beams tend to "wobble" around there trajectory and aside the dipole accelerator magnets, there are also quadrupole and sextupole magnets to keep the beams focussed and on track. Just search the net for quadrupole + lhc and you'll find some interesting). Getting off-topic here ... back to the LHC ...

In order to be able to get to higher energies, the LHC uses protons that are much heavier, hence at the same speed have more energy (E=mc²). But to accelerate these in two opposite directions, they need two opposite oriented magnetic fields, hence the two tubes needed.

PS: it is rather difficult to exactly "align" the proton bunches and let them collide at the right spot, and even then, you have to imagine the proton density as swarms of mosquito's flying in opposite directions through a giant tube: the chance two "protons" actually interact and create some interesting new particle is still "small", if it wasn't, after a few cycles, all protons would have all reacted with each other and the beams would have vanished.

PS2: now try not to flip: the "low density" of protons is still resulting in 40.000.000 collisions per second (just multiply the bunch density with the number of times bunches crossover with speed of light, a 27km tunnel is traveled many times per second !), and then realize every collision generates megabytes of data, so we are talking of petabytes of data each second. of course even with the grid it is impossible to store and process all this data, so the experiment will filter out of the 40.000.000 collisions a second the 100.000 collisions of data that show one of the predefined "interesting" signatures, and this in 25 nanoseconds time; the 100.000Hz can be studied for a full second or so and then only the 100 most interesting of these are kept and put on the grid.

Thats my understanding of what happens ... feel free to correct as I'm not a professional physicist at all (but an electric engineer with a big interest in physics and mathematics)

[jmayes]

Remember the LHC's predecessor, the LEP (Large Electron Positron collider) that was housed in the same tunnel, was running the two beams in the same (oval shaped) pipe. They could do this because the beams of electrons and positrons have different electric charges, so a magnetic field will force the particles in opposite directions. The size of the particle bunch is order of magnitude mm, the beam pipe diameter cm. Now the magnitude order for the area is squared ! So there is lots of room for two beams in one pipe without "hitting" each other (or the wall of the tube, as the beams tend to "wobble" around there trajectory and aside the dipole accelerator magnets, there are also quadrupole and sextupole magnets to keep the beams focussed and on track. Just search the net for quadrupole + lhc and you'll find some interesting). Getting off-topic here ... back to the LHC ...

Ok, I think that cleared this up for me- both beams are actually in the same tube as they pass through the experiments, I would surmise that when they want a collision they set the magnets to pull the beams together at just the right time.

Thankx for the help!
J

[chriwi]

hello jmayes,

that sounds like a rather rasonable picture of whats going on only that the right time might be rather a matter of minutes than small fractions of miliseconds, as far as I understand they will pull the two beams togather for a rather long time (at least several minutes) in wich collisions will occur whnever 2 protons happen to meet headon, the rest of the beam will just pass through and continue circulating.

[jmayes]

My continuing thought leads me to believe that the beams must be stacked vertically when in the experiments and not colliding as they have to cross before leaving the experiment *but* I was also under the impression that the beam is a series of bursts which would allow them to time things out to decide to collide or not.

So I guess it comes down to if the beams are permanently combined in the experiments OR are the beams pulled together when collisions are wanted.

Would be nice to get the definitive answer for us gear heads.


Thankx again all!
J

[Xymox]

i will ask a authority on the subject. Im sure the answer will get explained completely shortly.

[daavery]

in most of the ring the beams are side-by-side in 2 beam tubes. in the experiment areas the 2 tubes merge into a single tube. once the beams are "stable" the paths of the 2 beams are slowly steered so they cross at the desired crossing point in each experiment area. that continues as a steady state until the beams have depleted so much that they need to be dumped and reloaded. the colliding state is expected to last hours once everything is operational

[pcatom]

In the arcs of the LHC the beams run in separate vacuum pipes spaced by 194mm horizontally. Around each experiment there are long straight sections each around 500m long. In these regions there are special dipole magnets at either side of the experiment which steer the two beams into a single pipe (if you like names these are called D1 and D2) and apart again at the other side ...
B1 <- ________ .................._______ -> B2
........................\____X____/
B2-> _________/ ................\_______ <-B1

When we want to keep the beams from colliding we can use several smaller dipole correctors to make a separation bump across the interaction region. These are used to separate the 2 beams by a few mm. In Atlas and Alice the separation is Horizontal, in CMS and LHCb it is vertical. These bumps would normally be on for injection, acceleration and adjusting the beams and switched off just before going into collisions. At present we don't bother using them and always have collisions.

[Xymox]

There you go..

[RonnyV]

Didn't know the crossing of beams had a different orientation in Atlas/Alice versus CMS/LHCb, so then I ask myself: why ? Is it so "by accident" or is there a reason to use both horizontal and vertical ? and why not diagonal ?

[Xymox]

I would imagine that different orientations produce different particle spray patterns ? Or maybe the orientation is needed for the experiment. ?

[chriwi]

It probably also depends on the geometry of the experiments and the fact where is space to put the respective magnets for the crossover.