- #1
- 37,123
- 13,964
The LHC is about to restart after a break of two years - beams could go around as early as today (you can watch it here but that is very technical). There are some misconceptions about it and questions I got asked multiple times, so I thought I write some answers to clarify how "working for the LHC" works. This is the n+1th place where you can find those answers, but with a different focus and at a different place. Feel free to ask more questions!
With "working for the LHC", I actually mean "working for one of the experiments at LHC" - the LHC is just the accelerator, it has four big experiments (ATLAS, CMS, ALICE, LHCb) and some smaller ones."You work for a CERN experiment, but not at CERN?"
Right. Most of the work is data analysis in some way - the detectors produce huge amounts of data (roughly 1 TB per second) that have to get filtered (there is no way to store all of it) and analyzed. Data is stored in a worldwide computing grid and can be accessed from everywhere in the world. There is no need to be at CERN all the time. Detector development for future upgrades is done all over the world as well.
There are a few scientists that actually run the detectors from the control rooms at CERN in shifts, and a few more experts are available to come to the control rooms if necessary - but these are very small groups compared to the size of the collaborations (about one thousand for ALICE and LHCb, several thousands for ATLAS and CMS).
There are also scientists working on the actual detector hardware, exchanging parts and so on while the LHC is shut down, but again that is a relatively small group.
Many important meetings and workshops are at CERN, so visits there (typically for 1-2 weeks) are common. In general, meetings are very important to stay up to date - what is done where, who does what (to avoid doing the same thing multiple times, or not at all) and so on. You cannot visit CERN for every meeting (typically several per week that can be related to your work), so most of the time the meetings are done via the internet."When do they take your data?"
Many experimental sites work like that - you get some hours to weeks of time to run your experiment, and then you leave again and analyze what you got. The LHC does not work that way: it collides protons with protons (sometimes lead ions) and you cannot choose what happens in the collisions. The experiments have to record everything that could be interesting, every time the LHC is colliding protons. Ideally the LHC would run 24/7 but 30%-40% of the time is more realistic due to various technical issues. Analyses often look for rare processes or need a very high precision (you want to be better than the previous analysis), so they usually need the collision data from months to years to get enough statistics. Both ATLAS and CMS had about 2*1015 collisions so far, LHCb and ALICE have lower numbers."How do you collide protons?"
This is not as trivial as it sounds like. Protons are tiny - 10-15 m in diameter. There is no way to focus them well enough to let two specific protons collide. The LHC collects about 100 billion protons in groups called "bunches", thinner than a human hair and with a length of a few centimeters. In 2012 there were about 1400 of these bunches per direction circulating, the plan for 2015 is to have 2800. These bunches are "collided" - most protons just go through without any influence, but a few of them (~25-30 in ATLAS and CMS, ~3 in LHCb, 0 to 1 in ALICE) collide. The others go around the ring to have another chance of collision later. The bunches were 15 m apart (soon 7.5 m), giving collision processes 20 million times per second (soon 40 million)."Do you (personally) study the Higgs boson?"
No. The LHC experiments study many different particles, the Higgs is certainly an important one but not the only one.Event displays
When you read news articles about the LHC, they often include colorful pictures like this one. While it can be possible to make a guess what happened based on the picture, no one actually looks at them to study the collisions. There is no way to look at 1015 of these pictures, you need some way to let a computer analyze them. In addition, you are interested in numbers these pictures do not show.None of the particles we study hit the detectors
This is often surprising. All the particles that fly through the detector are well-studied and the detectors are not able to improve our knowledge about them. So how does it work?
The particles we are interested in are very short-living. They decay before they would reach even the innermost parts of the detector. Their decay products might decay again, but eventually you get particles that live long enough to fly through the detector, which records their particle type, flight direction and energy. Based on these decay products, it is possible to reconstruct what happened in the collision.
This is not completely true, a few particles are living long enough to have a chance to reach the detectors, but that is very uncommon. And even there you are mainly interested in finding their decay products instead of the actual particles.Is it possible to visit the detectors?
Yes! Unfortunately, the best time to do so was in the last two years during the long shutdown. The LHC does shorter breaks every winter and longer ones are planned for the future as well, there might be more chances.
More LHC articles:
Part 2: https://www.physicsforums.com/insights/lhc-part-2-commissioning-2/
Part 3: https://www.physicsforums.com/insights/lhc-part-3-protons-large-barn/
Part 4: https://www.physicsforums.com/insights/lhc-part-4-searching-new-particles-decays/
With "working for the LHC", I actually mean "working for one of the experiments at LHC" - the LHC is just the accelerator, it has four big experiments (ATLAS, CMS, ALICE, LHCb) and some smaller ones."You work for a CERN experiment, but not at CERN?"
Right. Most of the work is data analysis in some way - the detectors produce huge amounts of data (roughly 1 TB per second) that have to get filtered (there is no way to store all of it) and analyzed. Data is stored in a worldwide computing grid and can be accessed from everywhere in the world. There is no need to be at CERN all the time. Detector development for future upgrades is done all over the world as well.
There are a few scientists that actually run the detectors from the control rooms at CERN in shifts, and a few more experts are available to come to the control rooms if necessary - but these are very small groups compared to the size of the collaborations (about one thousand for ALICE and LHCb, several thousands for ATLAS and CMS).
There are also scientists working on the actual detector hardware, exchanging parts and so on while the LHC is shut down, but again that is a relatively small group.
Many important meetings and workshops are at CERN, so visits there (typically for 1-2 weeks) are common. In general, meetings are very important to stay up to date - what is done where, who does what (to avoid doing the same thing multiple times, or not at all) and so on. You cannot visit CERN for every meeting (typically several per week that can be related to your work), so most of the time the meetings are done via the internet."When do they take your data?"
Many experimental sites work like that - you get some hours to weeks of time to run your experiment, and then you leave again and analyze what you got. The LHC does not work that way: it collides protons with protons (sometimes lead ions) and you cannot choose what happens in the collisions. The experiments have to record everything that could be interesting, every time the LHC is colliding protons. Ideally the LHC would run 24/7 but 30%-40% of the time is more realistic due to various technical issues. Analyses often look for rare processes or need a very high precision (you want to be better than the previous analysis), so they usually need the collision data from months to years to get enough statistics. Both ATLAS and CMS had about 2*1015 collisions so far, LHCb and ALICE have lower numbers."How do you collide protons?"
This is not as trivial as it sounds like. Protons are tiny - 10-15 m in diameter. There is no way to focus them well enough to let two specific protons collide. The LHC collects about 100 billion protons in groups called "bunches", thinner than a human hair and with a length of a few centimeters. In 2012 there were about 1400 of these bunches per direction circulating, the plan for 2015 is to have 2800. These bunches are "collided" - most protons just go through without any influence, but a few of them (~25-30 in ATLAS and CMS, ~3 in LHCb, 0 to 1 in ALICE) collide. The others go around the ring to have another chance of collision later. The bunches were 15 m apart (soon 7.5 m), giving collision processes 20 million times per second (soon 40 million)."Do you (personally) study the Higgs boson?"
No. The LHC experiments study many different particles, the Higgs is certainly an important one but not the only one.Event displays
When you read news articles about the LHC, they often include colorful pictures like this one. While it can be possible to make a guess what happened based on the picture, no one actually looks at them to study the collisions. There is no way to look at 1015 of these pictures, you need some way to let a computer analyze them. In addition, you are interested in numbers these pictures do not show.None of the particles we study hit the detectors
This is often surprising. All the particles that fly through the detector are well-studied and the detectors are not able to improve our knowledge about them. So how does it work?
The particles we are interested in are very short-living. They decay before they would reach even the innermost parts of the detector. Their decay products might decay again, but eventually you get particles that live long enough to fly through the detector, which records their particle type, flight direction and energy. Based on these decay products, it is possible to reconstruct what happened in the collision.
This is not completely true, a few particles are living long enough to have a chance to reach the detectors, but that is very uncommon. And even there you are mainly interested in finding their decay products instead of the actual particles.Is it possible to visit the detectors?
Yes! Unfortunately, the best time to do so was in the last two years during the long shutdown. The LHC does shorter breaks every winter and longer ones are planned for the future as well, there might be more chances.
More LHC articles:
Part 2: https://www.physicsforums.com/insights/lhc-part-2-commissioning-2/
Part 3: https://www.physicsforums.com/insights/lhc-part-3-protons-large-barn/
Part 4: https://www.physicsforums.com/insights/lhc-part-4-searching-new-particles-decays/