 You probably know the large hydrogen collider over at CERN. We heard quite a bit of it in the recent talks. This time, we will have a deep dive into the infrastructure. You can assume over the next speakers are doing a great job. Basically, it's pretty obvious because we're not stuck into a giant supermassive black hole. So please welcome, with a very warm approach, Severin Ernst Stephan. Thanks. Thank you. Hello, everyone. Hello, everyone. Thank you for coming. Thank you for coming here. So many people in the room. In the last few years, we had a few talks here on the subject of LHC and the physics that is being examined there. So how physics researches the large data amounts, how they deal with large numbers. Three years ago, we had a talk by Axel about how physics analysis mass of the data. Three years ago, we had Alex's talk. And last year, we took him from Carsten and Stephan. And today, we want to get a little more into the engineers' performance. Stephan is an electro-technician who works in the experimental field. And it's all about the reliability of the digital components. I'm Severin. I also work at CERN. I deal with the machine security. A short disclaimer ahead of time. The LHC is very large and very complex. We try to describe it as well as possible. 45 minutes are basically not enough to describe everything. But we'll tell you something about the sub-systems that we trust, or that we have the most knowledge about. So imagine the accelerator. Imagine you wanted to build a speedometer. What doesn't work is the size of the LHC because it has 30 kilometers in diameter. But there are three basic challenges. First, you have to accelerate the parts. Second, you have to keep the parts on the circuit. And last, you have to make sure that the parts in the accelerator are not accelerating what they shouldn't. So the solution you found at the LHC is a acceleration mechanism that is based on radio frequencies. You have superconducting magnets and you have superconducting magnets. And in order not to get collisions that you don't want, the parts fly in the vacuum. A short overview of the CERN itself. Many of you know that the CERN is in the vicinity of Genf, in the southwestern part of Switzerland. If you zoom in a bit, you can see the red circle of the LHC which maybe serves as a perspective that up there is a airport with a runway length of three kilometers. That's the Genfa Se. And at the back you can see the Mont Blanc from the CERN. If you zoom in a bit more, you can see the very big collider of the LHC. A little bit smaller on the other side. And in the big one there are two campuses, one in France and one in Switzerland. There are additional points. If we go through it briefly, we have the first one. This is the Atlas Experiment. This is one of the big and the opposite side of the LHC. On the exact opposite side you have the CMS. A little bit smaller, that's Alice. That's built specifically for Blythe ions. Another one is the LHCB Experiment that is specifically for non-symmetric physics. That's probably what we know about it. There are other LHCF, for example. It's about the front strung particles that are formed in Atlas. There's the Totem Experiment, also forward strung. Sorry. We have Moodal, a two-dimensional experiment at the eighth point. Totem at the fifth point. And a very new experiment now called Phaser. And it's just being built and it's probably going to be in operation in 2020. Where it's about forward strung again, where it's supposed to be detected by neutrinos. In addition, the accelerators come in place themselves. And here's a beam dump, where the particles are dumped when they're supposed to come out of the ring. And in addition, it comes in at point three and seven, a point where simple generation service is done. So at point four there's the linear accelerators. They're relatively old. On the left side you can see the second generation. It's one of the older linear accelerators. They were used until last year. They're out now. And on the right side are the newer linear accelerators. The linear accelerators can better accelerate. Then we have the PSBs, the proton-synchrotron accelerators. You can see two pictures again. What's special about that is that there are four radiation rays. Then we have the proton-synchrotron accelerators. That's the next level of acceleration. It has only one radiation ray. And from there it goes to the super proton-synchrotron, the SPS. And it has a range of seven kilometers. And there we accelerate the particles the last time before they really fly into the LHC. Basically, you already know the different accelerators. CERN has actually even a few more. The most known is the CERN for the LHC-4. But there are a lot of them. There are also a lot of smaller accelerators before it goes to the big LHC. At the moment there is material science, antimatter and stuff like that. Everything is controlled in the CCC, that's on the north side. It looks like this. We have five CERNs that show each other and are always on 24-7. And there the accelerators, the entire energy, the cryonetic systems, and everything are monitored. Everything runs on Scientific Linux. That means it's our own Linux distribution that we use. And of course it's open source. So when we talk about the LHC radiation, we have two radiation. There is one with the clock and one against the clock. And they then collect at four controlled points. Most of the time in the year they fly protons, and then they are left on each other. And at the end of the year, from November to December, we do that with blyon. It's not really about a straight line of protons, it's more of a spaghetti of radiation. That's why there are always bunches. They are then collected. And with protons it's not that much, but with blyons it's up to 10 million at a time. In total we had up to 10 high 6 bunches a year. The LHC itself is 27 kilometers long. On this graph you can see a little bit of it. There are eight straight sections and eight curves. The curves are always there where we have a service area where maybe experiments take place. In this picture it's not so visible, so we have here one more exact thing. This is one of these straight sections where you can see the aluminum foil. There are no magnets around it. And you can see that in the corners there is a big tube around it. It's a very famous picture. Because you have these blue dipole magnets around it. This is a no-tunnel. It has a diameter of 3.8 meters and a range of about 26.65 kilometers. There is a tube in which all the cryogenic is running. So the cooling and the gas. Then we have the magnetic tube ourselves to turn the particles. And then the electrical installation to send electricity and signals to all the magnets. The geography and topography is a bit more complicated because we have a mountain area in the western part. And it's pretty hard. It's of course from limestone granite. And if you had made the whole thing straight, you would have had to dig deeper into the granite. That's why it was decided to put the RC tunnel a bit diagonally. With an increase of about 1.4 percent. The depth is between 50 meters at 0.8 and about 170 meters deep at 0.4. We have already talked a bit about magnets, but we would like to go into more detail. So why do we need magnets? Maybe you have already learned that if you have a magnetic field and loaded particles, you can bend these loaded particles around a bend in the magnetic field. Depending on the load of the particle, we bend it right or left. There are these famous right-hand and left-hand rules. And at the LHC we can't use normal magnets like normal magnets, but we have to use electromagnetic magnets because only these are fast enough to build an electromagnetic field to bend these fast particles enough. At the LHC we see dipole magnets with a field of 8.8 teslas and a current of over 11 amperes. We have two different magnets. We have the length magnets. These are the dipole magnets and we have the injection extraction magnets. These are also dipole magnets, but they are built a bit differently because the input and output must be very fast. They have to be started with a full magnetic field in a few microseconds. Then we have complete magnets and they have more poles and they are used to focus on the beam or to finally focus. In total we have 1200 dipole magnets, 850 quadrupole magnets and then about 4800 with higher degrees. The dipole magnets consist of two lenses. They are used to bend the beam and in the middle of the magnet you can see that there is a hole and then there is a metallic structure around it and next to the hole there are these two metal blocks that are formed around it. The magnet length is about 15 meters and the manufacture precision is about 1.5 millimeters. The four pole magnets are basically basically equally charged particles and then the particles that are like equally charged particles and then we could just hit the beam and we could maybe do something and then the particles that touch the edges and so we can use these four pole magnets to focus a beam or to finally focus. The quadrupole magnets mean that you have four magnets, four poles for these lenses. To go a little bit into detail about focusing and finally focusing there is a very normal unfocused particle current and that's why you go through the first quadrupole magnets and this is only done in one axis. That's a little bit of a problem. So we don't have any focus in one axis and in the other axis we have an end-focusing aspect and then we need a second focus magnet to focus the magnet even more. And here in the middle axis you can see how the beam looks like. Here you can see on the left that he has an unfocused beam and that he is then focused in one direction and then he is focused in the other direction and so you have to use several four pole magnets to effectively focus the beam. In the LFC magnets we have very high current we need this current because otherwise we wouldn't be able to bend the high energy beams and if we were to use normal cables it wouldn't be possible to build a magnet so we use materials that are called super lighters because they have a very good effect that they have almost zero resistance at a certain temperature point and if you reach this point or a lower temperature the current can flow through the material without any hindrance. For this we have to cool the magnets but what is difficult is possible and on the right you can see an old plot from 1911 in Denmark where a researcher named Heike Onnes examined the super light effect and found out what was the temperature needed for the super light. To show the difference between a normal cable and a cable you can see on the right that would be the amount of cable that you would need in a conventional copper to direct the same current as the superlighter that is up there and that's why you can see why we use superconducting cables. This material is basically used to operate a special core and we use 10 Kelvin and for our application we use minus 1.9 Kelvin. So. Sometimes the conditions of the magnets change and then there are small vibrations and sometimes there are too high currents and what happens is there is a quetch. We can see that because when there is a quetch the resistance of the magnets changes and we have a special electronic like this on the picture and with this system we measure the resistance at different places of the magnet and with an FPGA we measure it and then trigger an interlock which you can imagine and then there is a cable that goes through the entire cable there are thousands of switches and now that one of these systems is on then the entire machine so the entire speed ring is switched off and then of course the entire energy from the magnets and the electronics and for that there are the source resistance which you can see on the picture they are cooled in oil they are about 4 meters long and if it really comes to a quetch and through the resistance of the entire resistance then they heat up to 700 degrees Celsius and it takes a few hours until they are cooled out again and of course there are so yeah voltage voltage the problem with the power converters is that they to measure the current you have to in the right direction the current measurement must be very exact and have a very exact stability here you can see that in the 24 hours that they measure they can only have a deviation of 5 ppm that means 65 milliampere at 3000 13000 ampere current for that we had to build our own ADCs analog digital converters that can have this precision and the whole thing must also be placed in a very precise temperature controlled environment ok, cooling we have already told that we use super-cooled magnets and they have to be cooled our super-cooled magnets at the LHC have to be cooled down to 1.9 degrees Kelvin and when we start the LHC we cool down liquid nitrogen down to 6000 tons at a temperature of 80 Kelvin then we cool down the magnets down to 1.9 degrees Kelvin with the help of 140 tons of liquid helium to compare the space has a temperature of 2.7 degrees Kelvin that means the LHC is much cooler than space every magnet has to be cooled down for about a month and the 15 meter long magnets need to be shortened to about 1 millimeter at every of the 8 points of the LHC we have compressors that cool down the helium and when we have the compressed and cooled helium, we pump it down and with a helium it has about 4.5 Kelvin and down in the tube we have another compressor that then cools the helium down to 1.8 Kelvin so that the magnet can actually be cooled down and the interesting effect is that at about 2.17 Kelvin the helium superfluid that means that the helium can flow everywhere and it can even flow against the gravitation and that is very practical because we use the we use the liquid helium the advantage is that through the through the angle that the LHC has so it is straight in the mountain that the helium can actually flow up we sometimes have to pump the helium even against the gravitation and that is something we have to pay attention to because the LHC has such an angle that we basically get an idea of what we have for problems at the LHC and everything we have done was to deal with how we get the radiation through the radiation and now there is another problem namely the acceleration and with a big synchrotron like the LHC we want to make a short summary about the beam we said the LHC comes in bunches in packages with about 100 million protons and these bunches have a distance of about 25 nanoseconds when they fly past each other and the results are the tasks of the radio frequency system the bunches the packages of electrons are closely connected and the second is of course to accelerate they are directed into the accelerator with 450 giga electrons and then have to be accelerated in the accelerator you can now as a fast-changing electric field and these waves are directed into a so-called cavity and in there the field changes very quickly this RF signal that changes with the electrons with the bunches and with that the electrons the positive bunch are directed into the negative field from the positive and are tuned to the speed with which the bunches run through the LHC that means that the bunches that are too slow are a bit accelerated because they are too late and the bunches that are too fast are slowed down we call that longitudinal focus and that would be very inefficient if we would change the field for every bunch of electrons that's why we do 10 times that means the frequency is 10 times higher than the time to fly through the cavity and the bunches get much tighter and the focus is better here you see one of these cavities and how they are built in the tunnel 8 of these are used per strand and these are the things that are used to get the energy into the strand super light to reduce the thermal stress and they have to be built very precise and there is very small build tolerance so that everything is very exact the second thing you need is a source for the high power radio signal that's used to high power radio radio signal and each of those this clustron produces 300 kilowatts power and then you have a huge power because your radio has only 300 watts in total you have 16 of these clustrons that deliver a total power of 4.8 megawatts which are needed to accelerate the radiation in the NAC if we look at it and we solved one of the problems, namely keeping the particles focused and the frequency on a certain frequency then the particles stay together so if the particles are in the NAC then the frequency has to be increased slowly to increase the frequency the particles will be accelerated but this change is very very small sometimes it's less than one in one thousand and that's why the energy is slowly going up until the maximum energy of the particles is reached which is needed for the collision of particles now of course the question where does the signal that the frequencies that accelerates the timings the radio signal originally comes from a faraday kefi a very special scattered area and from there it is distributed to the lower level radio frequency for example the cavities and the clustrons and different methods to produce are used sometimes a proton bunch moves over a certain position of the LHC and then the other thing that you can observe is that there is a rotation clock which indicates that an orbit is filled so why is that now an important piece of information but when you look back at this picture that Serbine already showed about the accelerator complex the big complexity that all this creates is the synchronization of all these machines you just have to think that some of them fly around the LHC and then some of them are being added again and again and that's why the different accelerators have a synchronized time system to be able to synchronize the injection of new particles in the LHC so it looks like a time distribution system it's a very small section of it but it shows that there are huge spaces that are only there to distribute the right time to all parts of the LHC and the accelerators and if you think about it future colliders bigger colliders will need more time equipment and that's why Serbine has developed a time distribution system that is also open source and free for everyone so if you're interested but not only the accelerators are interested in this system but also other parts of the accelerators but also all experiments are interested that there is a certain time so you know why and especially when which collisions took place when the bunches flew past the experiment that they can control their tracking in a time axis and which events when the storage and processing pipeline took place so let's go to the acceleration and timing and now we come to the third problem and that is that the particles have to be protected so that they collide and we need the vacuum system again it's not as easy as a vacuum chamber but there is not only one vacuum system but there are three vacuum systems at the LHC the first two are maybe not so interesting the first are isolation vacuum systems they isolate especially thermally the magnets which have very low temperatures from the surrounding air to prevent them from heating up too much and exactly the same applies to helium tubes which are also isolated and the most important thing is the radiation vacuum that's probably the most interesting the tube in which the particles fly that is a picture of this tube where the particles fly and if you ask yourself if you want to make sure that there are no unwanted collisions how many molecules can be in there and typically a beam lifetime basically a beam lifetime a time of lightning so how long can you keep the particles in a circular run without them jumping up and destroying and as a time value the lightning should be running for about 100 hours without giving the requirement and that's why there are 100 femtobar air pressure in there to reach this level of vacuum you need many levels and many components to reach this vacuum so the original vacuum in the radiation tube that's running there has a total volume of about the Notre Dame church in Paris and we use turbomolecular pumps to get the air out and then you need more mechanisms because these pumps can't reduce the pressure and that's why we use a clever trick namely cryo pumps the outer tube of this radiation tube is where the extremely cold helium is pumped and what that does is that it leads to an effect called cryo pumps that means any molecule that hits the hill condenses there immediately and is then automatically removed from the atmosphere in this tube and increases the quality of the vacuum and with this help and some clever coatings of the tube you can reach the vacuum level that's needed to carry out these experiments a problem that's been solved here is that it creates a problem and that's the synchrotron radiation in the vacuum is an effect when you bend a part of the radiation then then it can happen that if an element hits the tube then there is no molecule and it goes back to the vacuum and therefore the vacuum gets worse another problem is that the synchrotron is also that there is a lot of heat and it creates a lot of heat and that leading over this helium is not that efficient to cool the helium is extremely energy efficient and therefore we need extremely much helium we need about 1 kilowatt of energy and we have another part that's called the beam screen the radiation and it also runs along the radiation and it heats the synchrotron of the heat and this tube is only cooled to about 20 Kelvin and that's much more efficient to cool the helium and the steel tube is about a millimeter thick and has pump holes where the hydrogen molecules can go through and inside it has a copper coating that is used to reduce the electric resistance which is necessary so that the radiation that would otherwise happen in this tube would increase and that's why there is so much engineering work in there so after we have talked about all the things that are necessary to bring a radiation we can now talk about the radiation and the instrumentalization of this radiation even without going yet even if we don't take care of the collisions the first thing you need is a part with which you can control the radiation and influence it that's called beam, so radiation cleaning you have to say that the radiation is never really super clean but it always travels with a halo of particles through the accelerators and these particles around the radiation and remove because they can produce these collisions in the magnets there are such slots that are built in two places in the LHC these are vertical and horizontal and you can switch them and that's why you can remove the particles that you want to remove and influence the beam and this can continue to circulate in the accelerators then there is another open topic when you operate a beam accelerator then you would like to know how the quality of the beam is and you need that to achieve the best performance because if you don't know what the current state is and the easiest measurement that you can do is the radiation it's something like the brightness, the intensity of the radiation and you want to know how much is flowing through the accelerators you use the same current to the same current converter and the particles that are flowing through create an induced voltage in the measuring device and you can measure the device is very big because you want to measure everything with this device from the smallest intensity in the beam to the highest performance with as many particles as possible to cover the luminosity the second is the luminosity measurement and the luminosity is the size with which you measure how much radiation is this measurement is used to optimize the injection rate for the target and for the interaction rate of the different experiments to match each other so it helps you to improve the statistics of your data and also you have to measure the angle in which the radiation is flowing and this angle is very small and that's why we have to measure it very precisely because it's less than a thousandth of a degree and that's pretty impressive because the principle of this measurement is pretty simple that's the second and the third would be the position measurement of the radiation because we want to always know where the radiation is at any time we have to measure it to influence it that's why you use the radiation position monitors you can see here these are capacitive sensors they have four points on the top left and on the top right they are distributed at a thousand places in the radiation and then they give a feedback to influence the radiation with the magnets you have a device that tells you where the radiation is and then you want to know how the intensity distribution is when you cut through the radiation where is the radiation strong and where not we use two different systems one is in X and Y once in the cross-section you can see through wire scanners very thin cables that run through and with this you can create a profile the other is the longitudinal profile along the acceleration axis and that means you can look if the radiation you can see the radiation when it runs through how does it look above that there is the beam loss monitor the radiation mostly all the magnets there is information about the radiation and there is also more than a thousand and the idea behind is that with more than a thousand that you have more than a thousand small interaction chambers where you put the particles and then you have the rain of the secondary particles you use that to protect the system because if there is too much loss of energy the system has to be turned down within a few nanoseconds because of that the accelerated safety because if the radiation would soften then it would destroy the entire acceleration and you would have to rebuild it and what we have talked about is the turning off of the LHC which sounds much easier than it is the last thing we want to talk about is the beam dam a trash can for the radiation and it consists of a cloth of about a ton of copper and if the beam is directed out of the accelerator then the entire energy of the radiation has to be distributed for that this device is particularly constructed there is a kicker magnet which can be turned on within a microsecond in a very short time and then controls the radiation in this copper block and directs the energy if that actually happens it heats up to 800°C and then has to cool down and whether you know or not currently the LHC is not the LHC is not in operation but it is switched off if it is switched on then there is this status which is available on all possible beam dams but also on the internet and there you can look at how the current state of the accelerator and the radiation can also be used to see if you as the operator of the radiation and where you have to go and who has the next layer to show you how it looks right now a black beam dam it is out of shutdown and inspections take place until 2021 you can see it from home because it is publicly available like most of what we do at the CERN and as I said it will not change much in the next year but it will be very interesting to follow how the commissioning of the system takes place when the shutdown is changed if you have the need to come to the CERN we can there is the possibility to visit the CERN the tours are free and you just have to visit the next website and then a appointment I would like to thank you for your attention and yes please make a queue and we start with a signal angel signal angel it is said to be a master red button it is said there was a big red button how often it must have been pressed and he switched it off like a shutdown button I could not understand it so well but I think the question was how often we use the beam dump system to delete the beam L if there is a emergency there are several emergency buttons almost every beam lockout switches the beam switch whenever we have the big hydrogen collider running then at the beginning we have a very high luminosity but after about 12 or 15 hours the luminosity goes down and down and then at some point the operator decides that he does not need the beam and that he will remove it and then he is completely removed and that happens sometimes every 12 hours sometimes every 24 hours something like that microphone 2 4 where does the energy come from do you have your own power we get all the electricity from the French network we get almost all our energy from there that's about 450 kilowatts and then we have our own transformers on site and a small part of the energy comes from the Swiss energy network but most of the time thank you for your question and microphone number one please thank you for your presentation thank you for the presentation the beam can warm if I understand you right then the beam can use a big block on 100 degrees on some 100 degrees heat could you use that as a weapon no, we don't believe we would never think about it and even if we knew it would probably be an impractical weapon but if the telephone is broken again then you can speed it up you said you can stop in nanoseconds you said you could stop within a nanosecond but the beam only moves 40 centimeters in a nanosecond how do you get that in such a short time no, what I said are the magnets that navigate the beam out of the LHC they have set times of just a few nanoseconds and then of course the particles have to have a complete rotation around the LHC to be guided do you have any photos of the entrance of the beam of the beam no, not really the picture that I showed was one of the few pictures that we could find the other one is not really open the LHC that started in 2008 and since then the beam dump system has not been opened because it is completely shot with a lot of stainless steel and that's why there are not so many pictures regarding power supply how do you switch the how do you switch the current do you switch the current or is there a special technology yes, the system was developed by CERN and we develop most of the technology that is so common with us but then for the production we often use the industry and yes, that are most of the classic current converters the interesting thing is that the current has to be measured and that's why we use these DCCTs that were mentioned before and all these control mechanisms that is the challenge you talked about the clock that runs on the left and on the right how can you determine what is the first bunch that is constructed so we don't have to build but it is a constructed idea somewhere here there is a counter that knows how many 40 megahertz clock cycles are a process and it just counts that is not measured with a real marker but there is a structure in the whole beam that is measured with the kicker magnets when they go up you could measure how the particles pass by the magnets but it is much easier to count the time you said you have very precise requirements on the timing clock now what is the question with the distance because of the light speed it is a consideration if we put a cable around the 27 km you have a big delay in the electric signal because of the signal and that has to be calibrated we have to measure that and then compensate for the signal need to arrive at the device is it dangerous to stand near the concrete blocks to the cooling in relation to radiation yes difficult yes it is not recommended it is not recommended the systems there are all very well closed doors with security systems it is very careful that that can not happen the access system is very well structured maybe you have watched the movie angel and demons there is something like that we have iris scanner on many doors where you have to scan iris so that people can get to the places where they are allowed there is a very thorough access control system what is the reason that the experiments are so far away on the other side basically you are talking about atlas and cms the reason is that when they were constructed there was a bit of fear that individual particles maybe interact from different experiments and therefore they were made as far apart as possible so that there would be no interaction between the experiments if i can see it correctly microphone number 5 yes hello i have seen that you are also using the canbus i have seen that you use the canbus why do you use it i know at least one use at least one use that is within an experiment it is thought about the complete control of all detectors on a canbus i don't know if there is actually an application within the accelerator but maybe if you come later you can find it do you have any official data about how many tons of data you have how many tons of gaffer tape in the whole thing no unfortunately we don't have what does it look like okay as far as can see the interwebs again with a question yes do you know your monthly power bill do you know the costs for the electricity no but it is i think the fact that what is known is that France it is actually the electricity is a donation that is one of the biggest contributions of france to the electricity and in winter the electricity is more expensive and that's why yes in winter the electricity is more expensive and that's why you switch the NAC for waiting times especially for winter because it is more expensive and that reduces the costs for the french contribution if we have dipole magnets then we bend the particles whenever we have dipole magnets in the arc we have to focus on the pattern and because we also have the same loaded parts in the bunch they will turn upside down and that's why we have to focus on them I don't see any questions no it doesn't look like that thank you Stefan and Severin a warm applause