0:00:20.230,0:00:21.230 36c3 preroll music 0:00:21.230,0:00:24.140 Herald: Our next talk's topic is the Large[br]Hadron Collider infrastructure talk. You 0:00:24.140,0:00:29.770 probably know the Large Hadron Collider[br]over at CERN. We heard quite a bit of it 0:00:29.770,0:00:37.080 in the recent talks. This time we will[br]have a deep dive into the infrastructure. 0:00:37.080,0:00:43.250 You can assume our next speakers are doing[br]a great job. Basically, it's pretty 0:00:43.250,0:00:48.360 obvious because we are not stucked into a[br]great into a giant supermassive black 0:00:48.360,0:00:55.530 hole. So please welcome with a very warm[br]applause, Severin and Stefan. 0:00:55.530,0:01:01.200 Applause 0:01:01.200,0:01:08.530 Severin: Yeah. Hello, everyone. Thanks for[br]coming. So many people here, quite nice. 0:01:08.530,0:01:12.580 In the last couple of years we had several[br]talks about, yeah, basically the physics 0:01:12.580,0:01:19.330 perspective of LHC, how physicists analyze[br]data at LHC, how physicists store all the 0:01:19.330,0:01:25.009 data, et cetera. And we would like to give[br]like more an engineering perspective of 0:01:25.009,0:01:30.039 the whole LHC. So three years ago we had a[br]talk by Axel about how physicists analyze 0:01:30.039,0:01:35.311 massive big data and then last year we had[br]a talk conquering large numbers at LHC by 0:01:35.311,0:01:40.369 Carsten and Stefanie. And we would, as[br]I've mentioned already, we would like to 0:01:40.369,0:01:47.069 give like more an engineering perspective.[br]We are Stefan and Severin. We're both 0:01:47.069,0:01:50.420 electrical engineers working at CERN.[br]Stefan is working in the experimental 0:01:50.420,0:01:54.929 physics microelectronics section and he[br]will give a second talk tomorrow about 0:01:54.929,0:01:59.700 designing high reliability digital[br]electronics together with Szymon tomorrow 0:01:59.700,0:02:05.349 morning at 11:30. And I'm, as I mentioned[br]already, also working at CERN. I'm working 0:02:05.349,0:02:10.119 in the electronics for machine protection[br]section. I will describe briefly later. A 0:02:10.119,0:02:17.629 short disclaimer; the LHC is a pretty big[br]machine and we try to explain it as good 0:02:17.629,0:02:21.870 as possible. 45 minutes is not really[br]enough to talk about everything because I 0:02:21.870,0:02:26.209 think you can basically take one of the[br]topics we are talking here about now and 0:02:26.209,0:02:31.069 talk for 45 minutes only about one[br]specific topic, but we try to give an 0:02:31.069,0:02:36.680 overview as good as possible. So imagine[br]you want to build an accelerator in your 0:02:36.680,0:02:41.540 backyard. OK, maybe not in your backyard[br]because LHC is quite big, so 27 kilometers 0:02:41.540,0:02:47.379 in diameter is quite big, but basically we[br]figured out three main challenges you have 0:02:47.379,0:02:50.680 to take. First of all, we have to[br]accelerate particles because otherwise 0:02:50.680,0:02:54.959 it's not a particle accelerator. Second,[br]we have to keep the particles on a 0:02:54.959,0:02:59.400 circular trajectory. And then third, we[br]have to make sure that the particles which 0:02:59.400,0:03:03.709 are inside our beam tube or beam pipe[br]don't collide with anything which is 0:03:03.709,0:03:08.939 there, for example, to beam pipe itself,[br]air molecules, etc. And the solution we 0:03:08.939,0:03:12.760 adopted for LHC there is, that we[br]accelerate the particles with a high power 0:03:12.760,0:03:16.980 radio frequency cavities. Then we have a[br]beam control system which is quite 0:03:16.980,0:03:22.010 sophisticated using superconducting[br]magnets and then we have the beam pipe 0:03:22.010,0:03:26.599 itself, which is evacuated, so it's under[br]vacuum conditions to avoid any collisions 0:03:26.599,0:03:33.069 we have inside with gas molecules, etc. A[br]brief overview about the location itself. 0:03:33.069,0:03:39.139 So probably many of you know already that[br]CERN is next to Geneva. So it's in the 0:03:39.139,0:03:43.079 western-southern part of Switzerland. When[br]we zoom in a little bit more, then we have 0:03:43.079,0:03:49.900 here an artificial like picture of LHC[br]itself in the red circle there. To put it 0:03:49.900,0:03:53.480 a little bit in a perspective, we have a[br]relatively big airport there. You can see 0:03:53.480,0:03:58.940 there, it's a 2200 metre long runway. We[br]have Geneva Lake next to it. And that's 0:03:58.940,0:04:03.469 only one small part of Geneva Lake, but[br]nevertheless, and what also quite nice, we 0:04:03.469,0:04:10.450 see Mont Blanc from LHC, er, from CERN.[br]When we zoom in a little bit more, then we 0:04:10.450,0:04:14.680 basically have the big, circular collider[br]there. That's LHC itself. And we have pre- 0:04:14.680,0:04:19.570 accelerators, I will talk in a few minutes[br]about. Basically we have two main 0:04:19.570,0:04:23.300 campuses: we have Meyrin Site, which is in[br]Switzerland, and we have to Prevessin 0:04:23.300,0:04:28.850 site, which is in France. Then at LHC[br]itself, we have eight service points. We 0:04:28.850,0:04:32.669 also call it just points, to briefly go[br]through all of them; we have point one 0:04:32.669,0:04:38.970 where we have the experiment called ATLAS,[br]one of the big and major experiments at 0:04:38.970,0:04:45.410 LHC. Then at the exactly opposite side of[br]ATLAS, we have CMS at point five. Then we 0:04:45.410,0:04:47.780 have a little bit smaller experiment,[br]which is ALICE. It was basically 0:04:47.780,0:04:51.950 constructed for lead ion runs. We will[br]talk about this later. And then we have 0:04:51.950,0:04:55.790 another relatively small experiment, it's[br]called LHCb. And that's the only non- 0:04:55.790,0:05:00.810 symmetrical experiment at LHC. These are,[br]I think, the four experiments you already 0:05:00.810,0:05:06.170 maybe heard of. Then there are four or[br]three other smaller experiments. We have 0:05:06.170,0:05:10.690 LHCf at point one, it's a forward[br]scattering experiment at point one. So 0:05:10.690,0:05:15.250 basically, they're taking data like[br]scattered particles from ATLAS itself. 0:05:15.250,0:05:19.871 Then we have TOTEM. It's also a forward[br]scattering experiment and point five, then 0:05:19.871,0:05:26.850 we have, sorry, we have MOEDAL, which is[br]the experiment at point eight. They're 0:05:26.850,0:05:33.260 looking for magnetic mono-poles. Then we[br]have TOTEM, sorry for that, at point five. 0:05:33.260,0:05:37.880 And then we have a relatively new[br]experiment which is called PHASER. It's 0:05:37.880,0:05:41.570 actually under construction and it will be[br]used, starting from 2021 and it's forward 0:05:41.570,0:05:47.720 scattering experiment, which, where they[br]try to detect neutrinos. Then we have 0:05:47.720,0:05:52.810 point four, there we have the RF cavities[br]to accelerate the particle beam itself. We 0:05:52.810,0:05:57.450 have the beam dump area. So when there is[br]like a fault in a machine or we just want 0:05:57.450,0:06:01.800 to dump the beam, then we used the mean[br]dump system at point six. And then we have 0:06:01.800,0:06:07.730 two more general service areas. It's point[br]three and point seven. LHC would not be 0:06:07.730,0:06:12.100 possible without the pre-accelerator[br]complex. So we have a relatively big one 0:06:12.100,0:06:17.350 and it's also sometimes relatively old. On[br]the left hand side of the slide you can 0:06:17.350,0:06:24.680 see LINAC2, it's an old linear accelerator[br]which was used until last year. It's not 0:06:24.680,0:06:27.920 now phased out. And now we have LINAC4,[br]which is also a linear accelerator and it 0:06:27.920,0:06:34.970 has a little bit higher acceleration. Then[br]we have the proton synchrotron booster. 0:06:34.970,0:06:40.310 It's the first circular collider. So you[br]can see two pictures there. What is 0:06:40.310,0:06:44.560 relatively special about PSB is that we[br]have there two, sorry, four beam pipes 0:06:44.560,0:06:49.970 instead of just one beam pipe. Then we[br]have the proton synchrotron accelerator, 0:06:49.970,0:06:56.620 which is the next stage for acceleration.[br]It then has only one one beam pipe. And 0:06:56.620,0:07:01.920 then we are going from PS we are going to[br]SPS, which is the super proton synchrotron, 0:07:01.920,0:07:07.740 which is has circumferences of seven[br]kilometers. There we basically accelerate 0:07:07.740,0:07:13.230 the particles the last time and then they[br]are injected it in the LHC itself. We 0:07:13.230,0:07:17.970 mentioned a few accelerators already,[br]basically all everything we just 0:07:17.970,0:07:23.040 highlighted here. But CERN is a little bit[br]more. So CERN is famous for LHC, I would 0:07:23.040,0:07:27.280 say. But there is much more than only the[br]LHC. So only about 15 percent of the 0:07:27.280,0:07:30.880 protons, which are accelerated in the pre-[br]accelerator complex, are really going to 0:07:30.880,0:07:36.620 the LHC itself. So there is much more:[br]there is material science, there is anti 0:07:36.620,0:07:43.460 matter research and all different other[br]kinds of research going on. Of course, 0:07:43.460,0:07:47.190 everything has to be controlled. It's[br]called a CCC, the CERN control center. 0:07:47.190,0:07:53.310 It's located at, the Prevessin site;[br]looks like that. Basically, we have, four 0:07:53.310,0:07:58.660 Cs looking to each other and there the[br]operators are sitting 24/7 and operate the 0:07:58.660,0:08:03.650 whole machine. So basically, the whole[br]pre-accelerator complex, all the energy 0:08:03.650,0:08:09.540 cryogenics and LHC itself. Before you ask,[br]everything is running on Scientific Linux. 0:08:09.540,0:08:13.770 So we have basically our own Linux[br]distributed distribution, which is used 0:08:13.770,0:08:19.280 there and it of course, it's open source.[br]Talking about the LHC beam itself, we have 0:08:19.280,0:08:22.830 two beams: one is running clockwise and[br]the other one is to running anti- 0:08:22.830,0:08:27.340 clockwise because we don't have a fixed[br]target experiment where we basically let 0:08:27.340,0:08:30.890 the accelerated particles colliding with[br]like a fixed target, like metal or 0:08:30.890,0:08:35.930 something like that. We have controlled[br]collisions at four points, we mentioned 0:08:35.930,0:08:42.180 before. Most of the year we have proton[br]runs. So we have protons and protons 0:08:42.180,0:08:46.520 colliding each towards each other. And[br]then we have at the end of the year, 0:08:46.520,0:08:52.280 nearly starting from November to December,[br]we have lead ion run. The protons itself 0:08:52.280,0:08:56.870 is not really like a fixed, straight line[br]of particles. We have something called 0:08:56.870,0:09:01.810 bunches. You can imagine a little bit like[br]spaghetti. It's basically the same length 0:09:01.810,0:09:06.949 of a spaghetti, but it is much thinner[br]than a spaghetti. And each bunch, when you 0:09:06.949,0:09:11.490 have a proton run, then each bunch[br]consists of approximately 100 billion 0:09:11.490,0:09:15.689 protons. And when you have lead ion runs,[br]then we have approximately 10 million lead 0:09:15.689,0:09:25.680 ions in LHC. And last year we operated[br]with 2565(sic) bunches in the LHC itself. 0:09:25.680,0:09:29.259 The LHC tunnel. We already talked about[br]the tunnel itself. It is 27 kilometers 0:09:29.259,0:09:33.839 long and you can see maybe a little bit on[br]this graph, that we have some, we have 0:09:33.839,0:09:38.920 eight straight sections and we have eight[br]arcs in the tunnel. Basically the straight 0:09:38.920,0:09:42.560 sections are always there, where we have[br]like service cavities or we have areas and 0:09:42.560,0:09:48.069 also the experiments. Because it's not so[br]good visible in this picture, I put the 0:09:48.069,0:09:51.700 picture here. Basically, that's a straight[br]section of LHC. You can basically just see 0:09:51.700,0:09:56.100 the beam pipe itself, with aluminum foil[br]around it, and there are also no magnets. 0:09:56.100,0:10:01.339 And when we look in the arc section of[br]LHC, then you see here the arc itself and 0:10:01.339,0:10:06.529 I think it's quite famous picture of LHC[br]itself because we have blue dipole magnets 0:10:06.529,0:10:12.509 there. The tunnel itself is an old tunnel,[br]used previously by LEP, the large electron 0:10:12.509,0:10:17.990 proton collider. It has a diameter of 3.8[br]metres and the circumference is 0:10:17.990,0:10:24.449 approximately 27 kilometers. Inside the[br]tunnel we have, first of all, cryogenics, 0:10:24.449,0:10:29.350 so we have big tubes, stainless steel[br]tubes to carry all the cryogenic. So 0:10:29.350,0:10:33.889 liquid helium and gaseous helium. Then we[br]have the magnet itself to bend the 0:10:33.889,0:10:38.809 particles and then we have electrical[br]installations to carry like signals from 0:10:38.809,0:10:43.779 the magnets to have safety systems,[br]electricity, etc, etc. Geography is a 0:10:43.779,0:10:49.829 little bit complicated in the area because[br]we have in the western part of LFC we have 0:10:49.829,0:10:55.029 the Jura mountain range and this Jura[br]mountain range has a relatively hard 0:10:55.029,0:11:00.190 material. It's not made out of, not made,[br]but nature. I mean it's a limestone, so 0:11:00.190,0:11:04.110 it's relatively complicated to dig into[br]this material, in comparison to all the 0:11:04.110,0:11:08.749 other areas at LHC. So when you would[br]basically put a straight section of LHC, 0:11:08.749,0:11:14.059 then you have to dig much more into the[br]relatively hard limestone. So that's why 0:11:14.059,0:11:18.480 it was decided that the LEP or LHC tunnel[br]is tilted a little bit. So we have a tilt 0:11:18.480,0:11:23.499 angle of one 1.4 percent there. The depth[br]is approximately between 50 meters at 0:11:23.499,0:11:32.000 point one or point eight, up to 170 metres[br]deep at point four. We already talked a 0:11:32.000,0:11:34.519 little bit about magnets, but we would[br]like to go a little bit more in the 0:11:34.519,0:11:40.639 details now. So why do we need magnets?[br]Um, maybe you learned at school that when 0:11:40.639,0:11:43.559 you have a magnetic field and you have[br]charged particles and you can bend 0:11:43.559,0:11:48.730 particles around an arc in a magnetic[br]field. Depending on the charge of the 0:11:48.730,0:11:52.009 particles, you bend them around on the[br]right side or left side, that's this 0:11:52.009,0:11:57.870 famous right hand and left side rule, you[br]maybe learned during school. And at LHC we 0:11:57.870,0:12:02.449 cannot use a normal magnets like typical[br]magnets. We have to use electromagnets 0:12:02.449,0:12:06.160 because normal magnets would not be strong[br]enough to build in an electromagnetic 0:12:06.160,0:12:11.759 field which is feasible to bend the[br]particles around the whole tunnel. At LHC 0:12:11.759,0:12:16.819 we use, in the dipole magnets, a magnetic[br]field of 8.3 Tesla. And to do this we need 0:12:16.819,0:12:24.329 a current of 11850 amps. We have basically[br]two different types of magnets. We have 0:12:24.329,0:12:29.050 bending magnets. So the dipole magnets I[br]mentioned quite often already. And then we 0:12:29.050,0:12:31.689 have injection and extraction magnets.[br]They are also dipole magnets there, but 0:12:31.689,0:12:35.620 they are a little bit differently[br]constructed, because the injection and 0:12:35.620,0:12:39.699 extraction magnets have to be quite fast,[br]because they have to basically be powered 0:12:39.699,0:12:47.350 up at full, the full magnetic field in[br]several microseconds. Then we have higher 0:12:47.350,0:12:50.660 order magnets which are quadrupole pool,[br]magnet, sextupole magnets and octupole 0:12:50.660,0:12:54.690 magnets, et cetera, et cetera. And they[br]are used for focusing and defocusing the 0:12:54.690,0:13:01.449 beam itself. In total, we have 1200 dipole[br]magnets at LHC. We have around 850 0:13:01.449,0:13:06.619 quadrupole magnets and we have 4800 higher[br]order magnets. But they are normally quite 0:13:06.619,0:13:12.559 short or so, shorter than the the other[br]magnets. The dipole magnets consist of two 0:13:12.559,0:13:20.060 apertures. They are used to bend to beam[br]around, so I already said. In the middle 0:13:20.060,0:13:24.519 of the magnet itself we have a cold bore.[br]So there is basically there are the 0:13:24.519,0:13:28.600 particles flying around. Then there is a[br]metallic structure. You can see this in 0:13:28.600,0:13:33.279 the picture. It is just a shiny metallic[br]sphere you see there. And then we have 0:13:33.279,0:13:38.720 next to the cold bore, we have the tool,[br]the two apertures to bend the particle and 0:13:38.720,0:13:43.850 build the magnetic field itself. The[br]dipole magnets have a length of 50 meters 0:13:43.850,0:13:47.889 and the manufacturing precision is plus[br]minus one fine, one point five percent, 0:13:47.889,0:13:53.660 er, one point five millimeters. Then we[br]have quadrupole magnets. They are used for 0:13:53.660,0:13:58.519 focusing and defocusing the beam. The[br]problem is that we have bunches, were are 0:13:58.519,0:14:04.440 basically equally charged particles inside.[br]And the Coloumb force tells us, that when 0:14:04.440,0:14:08.970 we have equally charged particles, then[br]they are basically want to fade out from 0:14:08.970,0:14:12.649 each other and in the end they would just[br]hit the beam pipe itself and we could 0:14:12.649,0:14:17.200 maybe destroy the beam or cannot do any[br]collisions. So what we do is we use a 0:14:17.200,0:14:22.470 quadruple magnets as, yeah, similar to[br]lenses, because we can focus and defocus 0:14:22.470,0:14:28.490 the beam. The quadrupole magnets, the name[br]already suggested it, that we have 0:14:28.490,0:14:32.380 basically four apertures. So we have on[br]the left and the right side two and then 0:14:32.380,0:14:38.459 we have on top and bottom we have also a[br]few of them. To go a little bit into 0:14:38.459,0:14:42.220 detail about the focusing and defocusing[br]scheme. In the beginning we have a 0:14:42.220,0:14:47.240 particle beam which is not focused, but we[br]want to focus it. Then we go to the first 0:14:47.240,0:14:53.679 quadrupole magnet. So we focus the beam.[br]And this is only done in one axis. That's 0:14:53.679,0:14:57.209 a little bit a problem. So, in the second[br]axis we don't have any focus and we have a 0:14:57.209,0:15:00.990 defocusing effect there. And then we have[br]to use a second quadrupole magnet for the 0:15:00.990,0:15:05.709 other axis, in this case the Y-axis to[br]focus the beam even further. And you can 0:15:05.709,0:15:10.800 even see this here in the Z-axis, that's[br]basically the cut off the beam itself. You 0:15:10.800,0:15:14.050 can also see that in the beginning we have[br]on the left side, we have an unfocused 0:15:14.050,0:15:17.929 beam and then we focus it in one axis, so[br]we have like a little bit more ellipse and 0:15:17.929,0:15:22.069 then we focus in the other direction, then[br]we have a different ellipse. So we have to 0:15:22.069,0:15:25.509 use several quadruple magnets in a row to[br]really focus the beam in the way we want 0:15:25.509,0:15:33.190 to have it. In the LHC magnets, we have[br]quite high currents. We we need these 0:15:33.190,0:15:39.179 currents, because otherwise we cannot bend[br]to the very high energetic particle beam 0:15:39.179,0:15:44.050 and to use normal conducting cable, it[br]would not be possible to basically build a 0:15:44.050,0:15:47.929 magnet out of it. So what we do is, we use[br]materials which are called superconducting 0:15:47.929,0:15:54.170 materials, because they're for very good[br]effect. They go to basically zero 0:15:54.170,0:16:00.410 resistance at a specific temperature[br]point. And after this point or when we 0:16:00.410,0:16:08.149 basically go lower, then the current can[br]flow without any losses inside of it. But 0:16:08.149,0:16:10.860 to reach the state, we have to cool the[br]magnets quite heavily, which is not so 0:16:10.860,0:16:16.889 easy, but it can be done. And on the right[br]side you basically see a very historic 0:16:16.889,0:16:22.429 plot. That was 1911 in Denmark, a[br]researcher called Heike Onnes detected for 0:16:22.429,0:16:27.220 the first time superconducting effect in[br]mercury. And it was detected at 4.19 0:16:27.220,0:16:32.509 Kelvin. To show you a little bit the[br]comparison between a normal conducting 0:16:32.509,0:16:37.189 cable and a superconducting cable, as we[br]put the picture here. So that is basically 0:16:37.189,0:16:43.529 the same amount of cable you need to use[br]to carry, the thirteen thousand amps and 0:16:43.529,0:16:47.509 to do the same or to transport the same[br]amount of energy we also can use a very 0:16:47.509,0:16:51.130 small superconducting cable and I think[br]it's quite obvious why we use here 0:16:51.130,0:16:57.639 superconducting cables. At the LHC we use[br]Niobium tin(sic) as material. And this 0:16:57.639,0:17:01.470 material basically goes into a[br]superconducting state at 10 Kelvin. But to 0:17:01.470,0:17:07.530 have a safe operation to LHC, we have to[br]cool it down at 1.9 Kelvin. 0:17:07.530,0:17:11.500 Superconducting magnets have some[br]benefits, but also some downsides, so 0:17:11.500,0:17:16.439 sometimes they change their state because[br]there are small rigid vibrations and the 0:17:16.439,0:17:20.650 magnet or the temperature's not precise[br]enough or the current is too high, then 0:17:20.650,0:17:25.240 they change that state and it's called[br]"Quench". And, we basically can detect a 0:17:25.240,0:17:29.570 Quench when we measure the voltage across[br]the magnet, because the resistance changes 0:17:29.570,0:17:33.519 at this point. So when there is a Quench[br]then the resistance changes quite rapidly, 0:17:33.519,0:17:38.330 in milliseconds and we can detect this[br]voltage rise with sophisticated 0:17:38.330,0:17:42.530 electronics. On the right side, you see a[br]board I'm working on. So basically here, 0:17:42.530,0:17:47.610 we have a measuring system to measure the[br]voltage across the magnet. And then we 0:17:47.610,0:17:54.340 have a detection logic implemented in FPGA[br]to basically send triggers out and open an 0:17:54.340,0:17:58.549 Interlock loop. Interlock loop is a system[br]at LHC. You can imagine that little bit 0:17:58.549,0:18:03.269 like a cable going around the whole tunnel[br]and there are thousands of switches around 0:18:03.269,0:18:08.510 this Interlock loop. And as soon as one of[br]the detection systems basically opens to 0:18:08.510,0:18:13.669 the interlock loop, then basically the[br]whole machine will be switched off. And 0:18:13.669,0:18:16.530 what means switched off is basically, that[br]we will power down the power converter, 0:18:16.530,0:18:20.039 but then the energy is still in the[br]superconducting magnet and it has to be 0:18:20.039,0:18:24.190 taken out of the superconducting magnet.[br]And therefore, we use dump resistors to 0:18:24.190,0:18:30.600 extract the energy. And here you can see a[br]picture of such a dump resistor. It's 0:18:30.600,0:18:34.960 quite big. It's in a stainless steel tube,[br]oil cooled. It's approximately three or 0:18:34.960,0:18:41.470 four meters long. And basically, when[br]there was a Quench, and the energy was 0:18:41.470,0:18:44.480 extracted via these resistors, the whole[br]resistor is heated up by several hundred 0:18:44.480,0:18:50.179 degrees and it needs several hours to cool[br]it down again. Power converters; the power 0:18:50.179,0:18:54.679 converters are used to power the magnet[br]itself. So they can produce a current of 0:18:54.679,0:19:01.110 approximately 13000 amps and a voltage of[br]plus minus 190 volts. And you can see a 0:19:01.110,0:19:06.310 picture how here, how big it is. One[br]downside with the power converters is that 0:19:06.310,0:19:10.269 they have to be, not downside but one[br]difficulty is, that they have to be very 0:19:10.269,0:19:16.210 precise, because every instability in the[br]current would have or has a direct effect 0:19:16.210,0:19:20.290 on the beam stability itself. So basically[br]the beam would be not diverted in the 0:19:20.290,0:19:27.169 right amount of length. So that's why they[br]have to be very precise and have to have a 0:19:27.169,0:19:32.409 very precise stability. So here I just[br]pointed out, like in 24 hours, the power 0:19:32.409,0:19:36.700 converter is only allowed to have a[br]deviation of 5 ppm. And in comparison, for 0:19:36.700,0:19:41.970 13000 amps we have a deviation of 65 milli[br]amps. So the power converters have to be 0:19:41.970,0:19:46.600 very precise. And to do that, we had to[br]develop our own ADC, because at the time 0:19:46.600,0:19:51.270 when LHC was built, there was no ADC on[br]the market which was able to have this 0:19:51.270,0:19:55.900 precision and also the whole ADC is put[br]into a super-precise temperature 0:19:55.900,0:20:02.970 controlled areas and it is calibrated[br]quite regularly. Okay, cryogenics. We 0:20:02.970,0:20:05.520 already talked about that we have[br]superconducting magnets and they have to 0:20:05.520,0:20:11.340 be cooled down quite low. So the[br]superconducting magnets we have at LHC has 0:20:11.340,0:20:17.230 have to be cooled down to 1.9 Kelvin. And[br]we are doing this when we like start the 0:20:17.230,0:20:21.610 LHC then we cool down on the first hand[br]with liquid nitrogen. So approximately six 0:20:21.610,0:20:26.429 thousand tonnes of liquid nitrogen are put [br]through the magnets to cool them down 0:20:26.429,0:20:33.260 to 18 Kelvin and afterwards we cool the[br]magnets down with liquid helium. And 0:20:33.260,0:20:38.960 liquid helium is at 1.9 or 1.8 Kelvin. And[br]to put it a little bit in a comparison, 0:20:38.960,0:20:42.200 outer space, so when we measure like the[br]temperature of space, we have 0:20:42.200,0:20:47.799 approximately 2.7 Kelvin in outer space.[br]So LHC is much colder than outer space. 0:20:47.799,0:20:52.289 The whole cooldown needs approximately one[br]month and each dipole magnet, which is 15 0:20:52.289,0:20:57.010 meters long, shrinks several centimeters[br]during that. Which also has to be taken 0:20:57.010,0:21:02.870 into account, because otherwise pipes[br]would break. The cryogenic system is that 0:21:02.870,0:21:08.980 we have at each of the eight points at LHC[br]we have compressors to cool down the 0:21:08.980,0:21:13.860 liquid helium or the helium itself. And[br]then we compress the helium and pump it 0:21:13.860,0:21:17.970 down. We have one gaseous helium stream,[br]which is at 15 Kelvin and we have liquid 0:21:17.970,0:21:24.490 helium stream at approximately 4.5 Kelvin.[br]And then we pump it underground and then 0:21:24.490,0:21:28.580 we have something called Cold Compression[br]System. And the Cold Compression System 0:21:28.580,0:21:36.090 even further reduces the pressure of the[br]helium that we have in the end a helium, 0:21:36.090,0:21:40.980 which is at 1.8 Kelvin. So it can really[br]cool down the magnet itself. And helium 0:21:40.980,0:21:45.600 has a very interesting effect because at[br]2.1 Kelvin, it becomes something called 0:21:45.600,0:21:52.350 superfluid. So it basically can run around[br]like holes, for example, or walls. It can 0:21:52.350,0:21:57.679 basically flow against gravity, which is[br]quite interesting. And it has also very 0:21:57.679,0:22:03.860 high thermal conductivity and that's also[br]why we use superfluid helium here. And 0:22:03.860,0:22:08.370 that's why we cool down the whole magnets[br]that low. And one other interesting effect 0:22:08.370,0:22:13.570 is also that the LHC tilt angle, which is[br]1.4 percent, has to be taken into account 0:22:13.570,0:22:18.320 because we have very low pressure inside[br]all the tubes or all the system, at 16 0:22:18.320,0:22:24.539 millibars. But we have sometimes to pump[br]the helium against gravity or going down. 0:22:24.539,0:22:28.481 So we also have to take into account the[br]LHC tilt angle to not have wrong pressure 0:22:28.481,0:22:34.519 levels at the whole LHC itself. Okay.[br]Stefan: All right! So, you probably 0:22:34.519,0:22:38.649 already got the idea, that what we've done[br]in the last 20 minutes, was only solve the 0:22:38.649,0:22:42.250 first of the three challenges we had,[br]which was actually bending the beam around 0:22:42.250,0:22:49.269 the circular trajectory. So I'm trying to[br]go to the other challenges we have lined 0:22:49.269,0:22:54.350 up in the beginning. And the first one of[br]that is the actual acceleration of the 0:22:54.350,0:22:59.700 particle beam. And large synchrotrons,[br]e.g. like the LHC, they use radio 0:22:59.700,0:23:05.299 frequency or RF systems to do this[br]acceleration. And I'm just going to do a 0:23:05.299,0:23:10.510 quick recap of the LHC beam and RF and how[br]they interact. So Severin mentioned 0:23:10.510,0:23:15.600 already that the particles in LHC actually[br]come in bunches. So in like packets that 0:23:15.600,0:23:20.309 contain about hundred billion protons and[br]those bunches are spaced when they are 0:23:20.309,0:23:26.470 running around the LHC approximately 25[br]nanoseconds apart. And starting from that 0:23:26.470,0:23:31.250 the tasks of the RF system are basically[br]twofold. It first has to ensure that these 0:23:31.250,0:23:35.630 bunches are kept tightly together in a[br]process that we call longitudinal 0:23:35.630,0:23:39.700 focusing. And the second task is to care[br]for the actual acceleration of the 0:23:39.700,0:23:44.570 particle bunches. So from their injection[br]energy, when they come from one of the 0:23:44.570,0:23:49.039 pre-accelerators up to their final energy,[br]that they are supposed to collide at 0:23:49.039,0:23:55.549 during the physics run. So in general, you[br]can imagine RF as being a quickly 0:23:55.549,0:24:03.260 alternating electric and magnetic field[br]components. And in the LHC, this RF energy 0:24:03.260,0:24:07.700 is basically injected into what is called[br]a cavity, which is a resonant structure. 0:24:07.700,0:24:11.520 And there the particle beams travels[br]through, while the field quickly 0:24:11.520,0:24:16.690 alternates and the RF signal, or the[br]energy, basically interacts with the 0:24:16.690,0:24:21.750 particle beam. So perhaps you know that[br]the protons are positively charged and 0:24:21.750,0:24:26.210 then a negative polarity of the field[br]would attract these protons, while the 0:24:26.210,0:24:32.320 positive field location would basically[br]move them away. And this has ... well, 0:24:32.320,0:24:36.260 after just injecting and with the[br]frequency of this RF field being the same 0:24:36.260,0:24:41.390 as the speed that the particles actually[br]go round the LHC, solves the first of the 0:24:41.390,0:24:45.110 two problems, which was the the focusing[br]because actually the particles that are 0:24:45.110,0:24:49.370 too slow arrive only when the field is[br]already changed to the opposite polarity 0:24:49.370,0:24:52.990 and actually get accelerated a bit, while[br]the particles that are too fast, they are 0:24:52.990,0:24:58.000 actually being decelerated a bit. And this[br]is a process that we call the longitudinal 0:24:58.000,0:25:03.490 focusing, which makes sure that the[br]bunches stay neatly packed together. And 0:25:03.490,0:25:06.090 of course this would be relatively[br]inefficient if we would only change the 0:25:06.090,0:25:12.399 polarity of this field once for each of[br]the proton bunches that pass by. Which is 0:25:12.399,0:25:15.870 why we do it ten times. So the polarity[br]basically changes ten times or the 0:25:15.870,0:25:21.370 frequency is ten times higher than the[br]bunch crossing frequency. And by doing 0:25:21.370,0:25:25.960 that, we make sure that the change of this[br]field is much faster and therefore the 0:25:25.960,0:25:33.070 particle bunches are packed much closer[br]together and the focusing is better. So 0:25:33.070,0:25:35.389 here you can see these cavities that were[br]shown in the previous picture as a 0:25:35.389,0:25:39.899 schematic, how they're actually placed in[br]the tunnel. So eight of these huge 0:25:39.899,0:25:43.850 cavities are used per beam and they are[br]the actual thing that is used to couple 0:25:43.850,0:25:49.730 the RF energy into the beam and transfer[br]it to the particles. They are also 0:25:49.730,0:25:54.179 operating superconductively, so at[br]cryogenic temperatures, to reduce the 0:25:54.179,0:25:59.610 thermal stress and the losses that would[br]otherwise occur in their materials. And 0:25:59.610,0:26:01.270 these are actually – even though they are[br]so big, similar to the magnets that had to 0:26:01.270,0:26:05.820 be very precisely manufactured – these[br]also have very small manufacturing 0:26:05.820,0:26:12.070 tolerances and have to be precisely tuned[br]to the RF frequency that is used to 0:26:12.070,0:26:17.440 inject. So and the second part of this,[br]that actually produces this high power RF 0:26:17.440,0:26:22.420 signal. For that is used what we call[br]Klystrons. So Klystrons are basically RF 0:26:22.420,0:26:29.330 amplifiers. They are built from high power[br]RF vacuum tubes and they amplify this 400 0:26:29.330,0:26:34.289 MHz signal that is used to transfer energy[br]to the particles. And each of those 0:26:34.289,0:26:40.100 Klystrons produces about 300 kW of power[br]and you can probably imagine how much that 0:26:40.100,0:26:43.889 power for an individual unit that is, if[br]you know that your microwave oven has like 0:26:43.889,0:26:49.460 2 or 3 kW. And of course, as we have eight[br]cavities per beam and one Klystron always 0:26:49.460,0:26:54.840 feeds one cavity, we in total have 16 of[br]those Klystrons and they are in principle 0:26:54.840,0:27:03.659 able to deliver a total energy of 4.8 MW[br]into the LHC beam to accelerate it. But if 0:27:03.659,0:27:06.640 we take a small step back for now, we have[br]only solved the first of the two problems, 0:27:06.640,0:27:12.510 which was to keep the bunches neatly[br]focused. Because currently the particles 0:27:12.510,0:27:17.429 have been injected and the frequency is at[br]some specific frequency and actually they 0:27:17.429,0:27:22.400 are only running basically in sync, the[br]two. So what we do after all the particle 0:27:22.400,0:27:26.410 bunches from the pre-accelerators have[br]been injected into LHC, is that we ever so 0:27:26.410,0:27:30.700 slightly increase the frequency, which of[br]course also means that the particles need 0:27:30.700,0:27:35.399 to accelerate together with the RF signal.[br]And this is the mechanism that we use to 0:27:35.399,0:27:40.159 accelerate them actually. And the change,[br]that is required to do this, is very tiny, 0:27:40.159,0:27:44.419 actually. So it is less than a thousandth[br]of a percent sometimes, that is used to 0:27:44.419,0:27:48.379 change the frequency to actually make them[br]go so much faster. So from their 0:27:48.379,0:27:54.070 relatively low injection energy up to the[br]top energy plateau that they need to have 0:27:54.070,0:28:00.409 to produce the actual physics collisions.[br]And an interesting question to ask here is 0:28:00.409,0:28:03.990 where does this signal actually comes from[br]if it needs to be so precisely tuned to 0:28:03.990,0:28:10.250 some specific frequency? Who generates it[br]or who controls it? And that opens up the 0:28:10.250,0:28:15.669 whole complex of the timing of the LHC, of[br]the machine. So actually this first signal 0:28:15.669,0:28:21.460 that I mentioned, this RF signal, it[br]originates in a Faraday cage. So an 0:28:21.460,0:28:25.880 especially shielded area somewhere on the[br]Prévessin site of CERN. And from there it 0:28:25.880,0:28:33.440 is distributed to the low-level RF[br]subsystem with the Klystrons and the 0:28:33.440,0:28:38.690 cavities. But inside this room, there are[br]also a number of other signals generated. 0:28:38.690,0:28:42.720 The first one of that being this Bunch[br]Crossing Clock, which is the actual clock 0:28:42.720,0:28:48.240 that signals one pulse, basically every[br]time, it changes polarity one time a 0:28:48.240,0:28:54.640 proton bunch moves across a specific[br]location inside the LHC. And another one 0:28:54.640,0:28:59.470 is the so-called orbit clock, which always[br]indicates the start of the first or when 0:28:59.470,0:29:04.990 one proton bunch has basically re-arrived[br]at the same position and has completed one 0:29:04.990,0:29:10.820 orbit. And you may ask the question why[br]this is an important piece of information. 0:29:10.820,0:29:16.529 But if you think back to this image that[br]Severin has already shown, about the 0:29:16.529,0:29:21.789 accelerator complex, the big challenge[br]that all this brings is also the whole 0:29:21.789,0:29:24.909 synchronization of all these machines.[br]Because you have to imagine that while 0:29:24.909,0:29:29.790 these proton bunches run around the LHC[br]and new ones are supposed to be injected 0:29:29.790,0:29:34.000 from the outside, from another pre-[br]accelerator, this has to be very precisely 0:29:34.000,0:29:37.840 synchronized. So all these pre-accelerator[br]systems actually share a common 0:29:37.840,0:29:42.840 synchronized timing system that allows[br]them to precisely inject a new packet of 0:29:42.840,0:29:49.380 bunches at the right position, at the[br]right location into the LHC. And this a 0:29:49.380,0:29:53.309 bit how such a timing distribution system[br]looks like. It is only a very small 0:29:53.309,0:29:56.430 excerpt of what it looks like, but it[br]gives you an idea that somewhere 0:29:56.430,0:30:01.100 underground in the LHC there is rooms full[br]of equipment that is just used to 0:30:01.100,0:30:06.110 distribute timing signals between[br]different parts of the accelerator. And of 0:30:06.110,0:30:11.909 course, as CERN is forward-thinking and[br]realized that future colliders will need 0:30:11.909,0:30:15.400 quite a bit more of all this[br]synchronization and that the requirements 0:30:15.400,0:30:20.050 for how precisely everything needs to be[br]synchronized is ever growing, they 0:30:20.050,0:30:23.300 actually developed their own timing[br]distribution standard which is also 0:30:23.300,0:30:28.270 openly available and available for[br]everybody to use. So if you're interested, 0:30:28.270,0:30:34.320 look that up. But of course, not only the[br]accelerator itself is interested in this 0:30:34.320,0:30:40.309 information about what particles are where[br]and how quickly they interact or how 0:30:40.309,0:30:45.050 quickly they go around. But also all the[br]experiments need this information, because 0:30:45.050,0:30:50.070 in the end they want to know "Okay, has a[br]collision occurred at some specific time 0:30:50.070,0:30:54.940 in my experiment?" and actually providing[br]this timing information about when bunches 0:30:54.940,0:30:59.789 have crossed their experiment locations is[br]also vital for them to really time tag all 0:30:59.789,0:31:06.149 their collision data and basically track[br]which bunches were responsible for what 0:31:06.149,0:31:11.559 kind of event or what event throughout[br]their whole signal storage and processing 0:31:11.559,0:31:17.120 chain, let's say. Good. So that is[br]basically challenge 2 out of the way. So 0:31:17.120,0:31:19.889 that was the acceleration of the actual[br]particles and all the associated issues 0:31:19.889,0:31:25.899 with timing. And the third issue we[br]mentioned was that the particles need to, 0:31:25.899,0:31:31.330 let's say, be kept from colliding with[br]anything but themselves or the other beam. 0:31:31.330,0:31:36.210 And that is what we, why we need vacuum[br]systems for. So, again, it is not as 0:31:36.210,0:31:41.200 simple as just putting a vacuum somewhere.[br]Of course not. Because in fact, there is 0:31:41.200,0:31:45.740 not only one vacuum system at LHC, but[br]there are three. So, the first two of 0:31:45.740,0:31:51.090 those are perhaps a bit less interesting[br]to most of us. They are mainly insulation 0:31:51.090,0:31:56.919 vacuum systems that are used for the[br]cryogenic magnets. So they isolate, 0:31:56.919,0:32:02.789 basically thermally isolate the magnets at[br]those very cool temperatures from the 0:32:02.789,0:32:08.789 surrounding air to avoid them getting more[br]heat load than they need to. And there is 0:32:08.789,0:32:11.830 an insulation vacuum also for the helium[br]distribution lines that are actually 0:32:11.830,0:32:16.299 distributing, delivering the helium to[br]these magnets. And then the third one, 0:32:16.299,0:32:19.820 which is perhaps the most interesting one,[br]is the beam vacuum. So the one where 0:32:19.820,0:32:24.999 actually the beam circulates inside the[br]LHC. And this is a cross section of what 0:32:24.999,0:32:30.242 this beam vacuum typically looks like. So[br]it is approximately this size, so a very 0:32:30.242,0:32:37.470 ... handful, let's say. And the question[br]you may ask "OK, if I want to keep all the 0:32:37.470,0:32:41.590 like the particles in my particle beam[br]from colliding with anything they are not 0:32:41.590,0:32:47.340 supposed to, for example, rest molecules[br]of remaining air there, how many molecules 0:32:47.340,0:32:51.240 can there still be?" So somebody has to[br]make up that number. And typically you 0:32:51.240,0:32:55.880 express this as a quantity called the[br]beam lifetime, which basically says if you 0:32:55.880,0:33:00.870 were only to keep those particles[br]circulating in the accelerator, how long 0:33:00.870,0:33:04.620 would it take until they have all[br]dispersed and lost their energy due to 0:33:04.620,0:33:10.129 colliding with rest gas molecules? And it[br]was decided that this should be at a value 0:33:10.129,0:33:15.230 of 100 hours, is what the beams should[br]basically be able to circulate without 0:33:15.230,0:33:19.340 collisions, without being lost. And this[br]gave the requirement for pressures down to 0:33:19.340,0:33:24.700 about 100 femtobar, which is a very small,[br]very, very tiny fraction of the 0:33:24.700,0:33:29.679 atmospheric pressure we have here, which[br]is about 1 bar. And to actually get to 0:33:29.679,0:33:34.019 this level of vacuum, it requires multiple[br]stages and multiple components to actually 0:33:34.019,0:33:42.399 get there. So the initial vacuum inside[br]these beam tubes, which are basically 0:33:42.399,0:33:48.610 going throughout the whole LHC tunnel, has[br]the volume of approximately the Notre-Dame 0:33:48.610,0:33:53.800 cathedral. So the first step of getting[br]all the air out of these beam tubes is 0:33:53.800,0:34:00.179 using turbomolecular pumps. And then there[br]needs to be more mechanisms to reduce the 0:34:00.179,0:34:04.000 pressure even further, because these pumps[br]are not able to reduce the pressure to the 0:34:04.000,0:34:09.231 levels required. And they actually use a[br]relatively clever trick to do that, which 0:34:09.231,0:34:16.169 is the use of cryopumping. So the, ... I[br]cannot show that? Okay. So the outer wall 0:34:16.169,0:34:20.380 of this beam pipe cross section that you[br]see here is actually also where the very 0:34:20.380,0:34:27.230 cold helium inside the magnets is outside[br]of. And what that does is, it leads to an 0:34:27.230,0:34:31.589 effect called cryopumping. So actually any[br]rest gas molecule that hits this wall 0:34:31.589,0:34:35.990 actually condenses there. And as the[br]molecules condense there, they are of 0:34:35.990,0:34:40.510 course removed from the atmosphere inside[br]this beam pipe, which removes them from 0:34:40.510,0:34:44.750 the atmosphere and increases the quality[br]of the vacuum. And with the use of this 0:34:44.750,0:34:47.760 and then the warm sections, the use of[br]getter coatings, which are basically able 0:34:47.760,0:34:53.609 to trap gas molecules, you are able to[br]reach the crazy vacuum levels that are 0:34:53.609,0:34:59.020 required to make this happen. But they[br]realized also during the design that one 0:34:59.020,0:35:05.430 big problem – for the first time in an[br]accelerator – another effect will create a 0:35:05.430,0:35:08.540 significant problem for the vacuum, which[br]is the generation of synchrotron 0:35:08.540,0:35:15.240 radiation. So synchrotron radiation is a[br]byproduct of when you do bend a particle 0:35:15.240,0:35:20.220 beam, it results in a phenomenon called[br]synchrotron radiation. And when this 0:35:20.220,0:35:24.369 synchrotron radiation, as it goes straight[br]on and is not bent, hits the walls of this 0:35:24.369,0:35:30.170 vacuum system, or in this case of the beam[br]pipe, it actually liberates molecules from 0:35:30.170,0:35:33.810 there and reintroduces them into the[br]vacuum, which of course then makes the 0:35:33.810,0:35:40.230 vacuum worse again. An additional problem[br]that gives the synchrotron radiation is, 0:35:40.230,0:35:44.960 that it also gives a significant heat[br]load, and if you need to dissipate all 0:35:44.960,0:35:49.660 this heat that is generated through the[br]very cold helium, this is not a very 0:35:49.660,0:35:53.820 efficient process. Because making this[br]helium so cool, is actually a very energy 0:35:53.820,0:35:58.590 intensive process. And just removing a[br]single watt of thermal power through the 0:35:58.590,0:36:03.230 superfluid helium costs about 1 kW of[br]energy. So that is not the most efficient 0:36:03.230,0:36:07.770 part. And this is why the cross-section[br]you have just seen includes another large 0:36:07.770,0:36:11.200 component, which also technically belongs[br]to the vacuum system, which is called the 0:36:11.200,0:36:15.940 beam screen. And this beam screen is[br]basically another tube running inside the 0:36:15.940,0:36:20.760 beam pipe, of which we have, of course,[br]two, which run inside the magnet cold 0:36:20.760,0:36:25.200 bores. And it shields the synchrotron[br]radiation heat load from the outer walls, 0:36:25.200,0:36:30.550 which are at 1.8 Kelvin, while this pipe[br]itself is actively cooled to only about 20 0:36:30.550,0:36:36.240 Kelvin of temperature, which is much more[br]efficient to dissipate this heat. So it is 0:36:36.240,0:36:39.440 basically a steel tube about one[br]millimeter thick. It has these pumping 0:36:39.440,0:36:46.920 holes, where hydrogen gas molecules can go[br]out of, and on the inside it has a copper 0:36:46.920,0:36:51.970 coating, which is used to reduce its[br]electrical resistance, which is required 0:36:51.970,0:36:55.530 because the beam, while it circulates,[br]also induces current that would otherwise 0:36:55.530,0:37:00.180 flow inside this tube, which is really, if[br]you think about it, only a simple tube and 0:37:00.180,0:37:03.950 it would increase the heat load again. So[br]a lot of engineering already has to go 0:37:03.950,0:37:11.579 into a very simple piece of ... a thing[br]like that. So after having spoken so much 0:37:11.579,0:37:16.450 about all the things required to just make[br]a beam circulate and accelerate and so on, 0:37:16.450,0:37:20.590 now it's probably also time to talk a[br]little bit about the beam itself and how 0:37:20.590,0:37:27.220 to control it and how to instrument, how[br]to measure things about this beam. Even 0:37:27.220,0:37:32.080 without going yet about collisions and[br]doing actual physics experiments. So the 0:37:32.080,0:37:36.760 first important bit that is able to[br]basically control or influence the beam 0:37:36.760,0:37:41.540 here is what's called the beam cleaning or[br]collimation system. So typically such a 0:37:41.540,0:37:47.050 particle beam is not very clean. It always[br]travels associated with what is called 0:37:47.050,0:37:51.640 halo of particles around this core area[br]that is less than a millimeter wide where 0:37:51.640,0:37:57.140 most of the intensity is focused. And[br]these particles outside we want to remove, 0:37:57.140,0:38:00.260 because they otherwise would be lost[br]inside the magnets and for example, would 0:38:00.260,0:38:05.640 lead to quenches of the superconducting[br]magnets. And for collimation, we basically 0:38:05.640,0:38:10.819 use small slits that are adjustable and[br]are located at two main locations of the 0:38:10.819,0:38:15.100 LHC. So they have collimation systems[br]there, with vertical and horizontal slits 0:38:15.100,0:38:21.590 that can be adjusted in width, in order to[br]scrape off all the particles that they do 0:38:21.590,0:38:25.319 want to get rid of and extract out of the[br]beam, while only the core part can 0:38:25.319,0:38:30.490 circulate and produce clean collisions[br]without any background, that otherwise 0:38:30.490,0:38:36.110 would need to be accounted for. And then[br]there is a whole other open topic of beam 0:38:36.110,0:38:39.710 instrumentation. So when you run a[br]particle accelerator, you want to measure 0:38:39.710,0:38:45.220 various quantities and performance figures[br]of such a beam. And that is crucial for a 0:38:45.220,0:38:48.730 correct operation and for the highest[br]performance, getting the highest 0:38:48.730,0:38:52.331 performance from an accelerator. And there[br]are a lot of different types of those, and 0:38:52.331,0:38:58.920 I want to go quickly about ... over why we[br]have them and what we do with them. So the 0:38:58.920,0:39:03.730 first and most basic measurement you want[br]to do, is the beam current measurement. So 0:39:03.730,0:39:08.550 the beam current is a basic accelerator[br]beam intensity measurement. So it gives 0:39:08.550,0:39:13.260 you an idea of how strong the beam that is[br]running inside your accelerator is. And it 0:39:13.260,0:39:18.000 is measured using these DCCTs or DC[br]current transformers. And their basic 0:39:18.000,0:39:22.230 principle of operation is, that while the[br]particles move through this torus, which 0:39:22.230,0:39:26.069 is actually a coil or a transformer,[br]induces a voltage there that you can 0:39:26.069,0:39:30.579 measure and then use to quantify the[br]intensity of this beam. And the big 0:39:30.579,0:39:34.760 challenge here is that the dynamic range,[br]this instrument needs to capture, is 0:39:34.760,0:39:39.930 really large, because it has to operate[br]from the lowest intensity pilot injection 0:39:39.930,0:39:46.119 beams up to the full energy, full number[br]of bunches running inside the LHC. So it 0:39:46.119,0:39:51.580 has to cover six orders of magnitude of[br]measurement dynamic range. Then the second 0:39:51.580,0:39:57.130 thing when talking about collisions is the[br]luminosity measurement. So luminosity is a 0:39:57.130,0:40:02.210 quantity basically said to measure the[br]rate of interaction of the particle beams. 0:40:02.210,0:40:06.710 So to give you an idea of how often[br]interactions happen inside the experiments 0:40:06.710,0:40:11.280 or where you want them to happen. And this[br]measurement is used to first of all, 0:40:11.280,0:40:14.461 adjust this interaction rate to a target[br]value, which is optimal for the 0:40:14.461,0:40:19.109 experiments to function and to equalize[br]the interaction rates in different 0:40:19.109,0:40:23.970 experiments. So different experiments also[br]are specified to have the same interaction 0:40:23.970,0:40:29.650 rate so they can get the same let's say[br]statistical quality of their data. So it's 0:40:29.650,0:40:33.869 used to equalize those. And then as a[br]third thing, this system is also used to 0:40:33.869,0:40:37.510 measure the crossing angle of the beam. So[br]as you may know, at some point, when the 0:40:37.510,0:40:42.610 beams are collided, they collide at an[br]angle, that is very small. And this angle 0:40:42.610,0:40:47.559 is actually measured also very precisely[br]in order to adjust it correctly. And it is 0:40:47.559,0:40:50.190 measured to less than a thousandth of a[br]degree, which is again a very impressive 0:40:50.190,0:40:54.640 feat, given that the detection principle[br]of this measurement is only measurement of 0:40:54.640,0:40:58.700 some neutral particles that are a result[br]of the particle interaction of the beam 0:40:58.700,0:41:05.080 ... of the collision. Okay, so that is[br]number two. Then number three that we have 0:41:05.080,0:41:09.610 is the beam position monitor. Because[br]along the LHC, you also always want to 0:41:09.610,0:41:14.940 know, where the beam is at any given time.[br]So you want to measure the position of the 0:41:14.940,0:41:19.190 beam inside the beam pipe in order to[br]optimally adjust it to the position you 0:41:19.190,0:41:23.740 want to have it. And for that we use these[br]beam position monitors of which we have 0:41:23.740,0:41:27.910 more than a thousand installed along the[br]LHC. So they are typically capacitive 0:41:27.910,0:41:31.780 probes or electromagnetic strip lines. As[br]you can see on top and bottom 0:41:31.780,0:41:36.790 respectively. And they basically are[br]distributed along the LHC and provide 0:41:36.790,0:41:40.300 position of the particle beam along the[br]accelerator, which can then be used to 0:41:40.300,0:41:47.640 tune, for example, the magnets. All right.[br]Then we have beam profile. So after the 0:41:47.640,0:41:51.059 position, that gives you an idea where the[br]beam is, you also want to know its 0:41:51.059,0:41:57.329 intensity distribution. Basically when you[br]do a cut through the beam pipe somewhere, 0:41:57.329,0:42:02.480 you want to know how the intensity profile[br]looks like. And for that we have basically 0:42:02.480,0:42:07.380 two measurement systems. One measures the[br]profile in X and Y directions. So if you 0:42:07.380,0:42:10.400 really would do a cut and it gives you[br]something like this and it's, for example, 0:42:10.400,0:42:14.950 done with wire scanners, which is literal,[br]very thin wire that is moved through the 0:42:14.950,0:42:20.290 beam. And then the current that the beam[br]moving through this wire, generates is 0:42:20.290,0:42:25.280 used to generate such a profile map, when[br]scanning with this wire. The other one is 0:42:25.280,0:42:29.750 the longitudinal profile, which gives you[br]an idea about the quality of your RF 0:42:29.750,0:42:34.040 system and there you want to know how the[br]intensity profile of your beam looks like. 0:42:34.040,0:42:37.690 If you were looking only at one spot of[br]the accelerator and the beam would pass 0:42:37.690,0:42:43.230 by, and you would basically see over time[br]how the intensity looks like. And then as 0:42:43.230,0:42:47.980 the last bit of beam instrumentation,[br]there is the beam loss monitors. So they 0:42:47.980,0:42:53.170 are these yellow tubes, that are located[br]on the outside of mostly all the magnets, 0:42:53.170,0:42:57.951 of the dipole and quadrupole magnets and[br]so on. Again, there's more than a thousand 0:42:57.951,0:43:03.230 of those. And the idea here is that you[br]need a lot of detectors that are basically 0:43:03.230,0:43:07.819 small ionization chambers, which detect[br]any showers of secondary particles that 0:43:07.819,0:43:12.720 are generated when one of the high energy[br]protons are lost somewhere in the magnet 0:43:12.720,0:43:16.760 materials. So these are really used for[br]protection of the system, because if a 0:43:16.760,0:43:21.109 specific threshold of energy loss is[br]detected, then the accelerator needs to be 0:43:21.109,0:43:26.220 quickly shut down. Which is why they have[br]to react in a matter of nanoseconds in 0:43:26.220,0:43:31.260 order to keep the accelerator safe.[br]Because any interaction of the particle 0:43:31.260,0:43:36.059 beam with for example, the magnets could[br]just destroy huge amounts of money and of 0:43:36.059,0:43:41.870 time that would be needed to rebuild. And[br]as a last and final thing, we have spoken 0:43:41.870,0:43:46.260 one or two times already about shutting[br]down the LHC. Which sounds also trivial at 0:43:46.260,0:43:51.799 first, but really is not. So, the last[br]thing here is, what we call the Beam Dump. 0:43:51.799,0:43:56.589 So the energy content that is contained in[br]those particular beams, it can be used, 0:43:56.589,0:44:02.950 could be used if it were shot on a copper[br]target, it could just melt 1000 kilograms 0:44:02.950,0:44:07.599 or one ton of copper instantly. So during[br]beam extraction, the process of getting 0:44:07.599,0:44:12.359 the particle beam outside out of the LHC,[br]this energy needs to be dissipated 0:44:12.359,0:44:16.990 somehow. And for that, this special Beam[br]Dump area is constructed. So there are 0:44:16.990,0:44:20.670 fast kicker magnets, that are used to ...,[br]that are able to ramp up in a really, 0:44:20.670,0:44:24.930 really short amount of time of[br]microseconds. And then the beam is 0:44:24.930,0:44:30.309 carefully and in a controlled manner[br]directed into a set of concrete blocks, 0:44:30.309,0:44:35.339 that is basically big enough to dissipate[br]all this energy, when required. And in the 0:44:35.339,0:44:38.800 process of doing so, it also heats up to[br]about 800 degrees Celsius, and then of 0:44:38.800,0:44:45.530 course, also needs the associated time to[br]cool down again. Good. So as you may or 0:44:45.530,0:44:50.040 may not know, currently the LHC is not in[br]operation. So LHC currently is undergoing 0:44:50.040,0:44:56.099 its second long shutdown phase, or LS2.[br]But what we do when the LHC is in 0:44:56.099,0:44:59.961 operation, is that we have these status[br]dashboards, that you can see here, that 0:44:59.961,0:45:04.970 are distributed all around CERN, and can[br]be used by anyone, any passer-by, to 0:45:04.970,0:45:09.980 basically monitor what the current[br]operation mode or the current situation of 0:45:09.980,0:45:15.609 the accelerator is. And can be used also[br]to quickly see if like an operator needs 0:45:15.609,0:45:19.990 to go somewhere or is needed, or how the[br]shift planning for the next shift works 0:45:19.990,0:45:24.430 out and so on. And on the right side you[br]would see what this currently looks like. 0:45:24.430,0:45:30.990 So basically black screen saying next beam[br]expected in spring 2021. And the good 0:45:30.990,0:45:33.980 thing about these status pages is that you[br]can actually see them from your home, 0:45:33.980,0:45:40.390 because they're also openly available, as[br]most of the stuff we do at CERN. So if you 0:45:40.390,0:45:44.510 are interested, then perhaps in a year[br]from now or a bit longer than a year, it 0:45:44.510,0:45:47.670 would be quite interesting to follow all[br]the commissioning process of when they are 0:45:47.670,0:45:54.130 trying to start the LHC back up, and[br]follow that process from your home. 0:45:54.130,0:45:58.109 Otherwise, if you now feel the urge to[br]maybe visit CERN, pay some of the things 0:45:58.109,0:46:01.980 we talked about a visit, or are just[br]generally interested, CERN offers a 0:46:01.980,0:46:06.970 variety of tours free of charge. So if[br]you're interested in that, visit that web 0:46:06.970,0:46:10.819 site and we would be happy to welcome you[br]there. And with that, thank you very much 0:46:10.819,0:46:14.569 for your attention. 0:46:14.569,0:46:17.530 Applause 0:46:17.530,0:46:24.470 Severin: Punktlandung.[br]Herald: Thank you, Stefan and Severin. If 0:46:24.470,0:46:29.869 you have questions, there are six[br]microphones in the room. Please make a 0:46:29.869,0:46:34.660 queue, and we start with the Signal Angel.[br]Signal Angel, please, first question. 0:46:34.660,0:46:39.630 Signal Angel: There is said to be a master[br]red button for shutting down the whole 0:46:39.630,0:46:45.859 system in case of heavy problems. How[br]often did you push it yet? 0:46:45.859,0:46:49.059 Stefan: Master red button?[br]Severin: Master button ... 0:46:49.059,0:46:55.160 Signal Angel: Like a shut down button.[br]Severin: I cannot really understand you. I 0:46:55.160,0:46:59.270 think the question was about how often[br]basically we used the Beam Dump system to 0:46:59.270,0:47:01.050 basically get rid of the beam, is that[br]correct? 0:47:01.050,0:47:04.280 Signal Angel: I guess so.[br]Stefan: He said there is a master button. 0:47:04.280,0:47:06.280 Signal Angel: I guess so.[br]Stefan: I think there's a master button in 0:47:06.280,0:47:08.280 the...[br]Severin: There is not only one master 0:47:08.280,0:47:11.790 button, there are several master buttons.[br]These are switches, called beam interlock 0:47:11.790,0:47:17.530 switch. Basically, at every operator's[br]screen, there is basically one beam 0:47:17.530,0:47:23.690 interlock switch. I don't know. I think[br]sometimes they get rid of the beam just 0:47:23.690,0:47:29.579 because, I mean. When we have LHC at full[br]operation – Stefan talked about the 0:47:29.579,0:47:33.799 luminosity – so what is happening, that in[br]the beginning we have a very high amount 0:47:33.799,0:47:39.920 of luminosity, So many particles collide[br]on each other. But over time, like after 0:47:39.920,0:47:45.230 12 or 15 hours or whatever, basically the[br]luminosity ..., so the amount of particles 0:47:45.230,0:47:49.119 which collide with each other, is going[br]down and down. So the luminosity 0:47:49.119,0:47:53.770 decreases. And then at some point in time,[br]basically the operators decide, that they 0:47:53.770,0:47:58.950 will now get rid of the actual beam, which[br]is inside LHC and basically will recover 0:47:58.950,0:48:02.290 the whole machine and then restart the[br]machine again. And this is done sometimes, 0:48:02.290,0:48:08.170 I don't know, every 12 hours, sometimes[br]after 24 hours. Something like that, yes. 0:48:08.170,0:48:11.380 Herald: Cool. And microphone number four,[br]I think. 0:48:11.380,0:48:17.559 Q: Yes. So where's the energy coming from?[br]So do you have your own power plant, or 0:48:17.559,0:48:20.119 so?[br]Severin: So, no, not really, not really. 0:48:20.119,0:48:24.349 Basically, we get all the power from the[br]French grid. So we have relatively big 0:48:24.349,0:48:32.640 power trails coming from the French grid.[br]So we get 450 kV of power there. So 0:48:32.640,0:48:35.390 basically the voltage is quite high and[br]then we have our own transformers on site. 0:48:35.390,0:48:39.609 And I think only, ... a little bit smaller[br]fraction of the energy is coming from the 0:48:39.609,0:48:43.940 Swiss grid. So basically we use most of[br]the energy which is coming from the French 0:48:43.940,0:48:46.940 grid.[br]Q: Okay. Thank you. 0:48:46.940,0:48:48.940 Herald: Thank you for your question. And[br]microphone number one, please. 0:48:48.940,0:48:57.280 Q: Hi. Thank you for your presentation. If[br]I'm not wrong, you say the beam can warm a 0:48:57.280,0:49:04.599 block of concrete to 800 Celsius. Would it[br]be possible to use it as a weapon? 0:49:04.599,0:49:09.960 Stefan: laughs Very likely not. And CERN[br]very much condemns these actions in any 0:49:09.960,0:49:13.690 form, I guess. So CERN operates in a[br]purely peaceful mission and would never 0:49:13.690,0:49:18.000 think about using their particle beams as[br]a weapon. And even if they could, it is 0:49:18.000,0:49:21.720 probably not the most practical thing to[br]do, I guess. laughs 0:49:21.720,0:49:26.710 Herald: But if your telephone is again[br]hanging up, you can destroy it, right? 0:49:26.710,0:49:29.090 Stefan: laughs[br]Herald: And microphone number six, I 0:49:29.090,0:49:33.390 think.[br]Q: Yes. So you said, you can stop in 0:49:33.390,0:49:39.980 nanoseconds, but just the light would go[br]just 30 centimeters, you know, a 0:49:39.980,0:49:45.110 nanosecond. How will you be able to stop[br]in this small time? 0:49:45.110,0:49:49.220 Stefan: Ah, no, no. So what I was talking[br]about is that these magnets that are used 0:49:49.220,0:49:54.770 to extract the beam out of the LHC, they[br]have reaction times, or ramp up times that 0:49:54.770,0:50:01.020 are in the order of 1, 2, 3 microseconds.[br]So not nanoseconds, but microseconds. And 0:50:01.020,0:50:06.329 really only then basically the particles[br]still circulate, worst case one full turn, 0:50:06.329,0:50:10.430 and only then moving outside of the[br]accelerator. 0:50:10.430,0:50:17.349 Herald: And microphone number one again.[br]Q: So do you have any photos of the front 0:50:17.349,0:50:22.609 of the dump block? It has to look like[br]it's got hit a lot. 0:50:22.609,0:50:26.110 laughter[br]Severin: No, not really. I think it's one 0:50:26.110,0:50:31.819 of the only pictures we could find about,[br]the Beam Dump system. And these areas, I 0:50:31.819,0:50:37.059 think it's not really opened any more. So[br]since operation of LHC, which was in 0:50:37.059,0:50:43.329 basically LHC started in 2008, and since[br]then, the Beam Dump system was not opened 0:50:43.329,0:50:48.860 again because it's completely sealed in[br]stainless steel. And that's why it wasn't 0:50:48.860,0:50:52.339 opened anymore.[br]Heral: Cool. Question from the interwebs. 0:50:52.339,0:50:59.329 Signal: Regarding power supply. How do you[br]switch or fine-control the currents? Are 0:50:59.329,0:51:03.460 you using classic silicone transistors,[br]off-the-shelf IGBTs? 0:51:03.460,0:51:07.230 Sverin: Um, yes. laughs[br]laughter 0:51:07.230,0:51:12.530 Severin: Yes. Uh, the system was developed[br]at CERN. And I think that's quite common 0:51:12.530,0:51:16.280 at CERN that we basically developed all[br]the technology at CERN or try to develop 0:51:16.280,0:51:20.250 nearly everything at CERN. But then[br]production, for example, is put into 0:51:20.250,0:51:25.750 industry. And yes, these are relatively[br]classical power converters. The 0:51:25.750,0:51:29.609 interesting or like challenging part about[br]the current power converters is really 0:51:29.609,0:51:32.859 that the current has to be measured quite[br]precisely and also controlled quite 0:51:32.859,0:51:39.430 precisely so there we use also DCC TS.[br]Which we have also mentioned before. But 0:51:39.430,0:51:42.290 basically all this controlled mechanism[br]there. That's one of the big challenges 0:51:42.290,0:51:47.980 there.[br]Herald: Cool. Microphone number one again. 0:51:47.980,0:51:55.609 Q: You talked about the orbit clock that[br]detects when the bunch is completed one 0:51:55.609,0:52:00.200 round. How is it possible to detect which[br]is the first bunch? 0:52:00.200,0:52:03.980 Stefan: Yeah. So it is it is actually not[br]detected, but this clock is actually 0:52:03.980,0:52:07.470 something that is constructed. So we[br]basically what we do is, we count these 0:52:07.470,0:52:13.250 cycles of the of the RF cycle. Maybe I can[br]open this slide. So somewhere there is a 0:52:13.250,0:52:18.510 counter that basically knows how many 40[br]MHz clock cycles a full rotation takes. 0:52:18.510,0:52:21.770 And then at some point decides this is[br]number one. And that's also where they 0:52:21.770,0:52:26.180 start counting when they inject bunches[br]into the LHC. So there's no marker, let's 0:52:26.180,0:52:29.510 say. But there is a certain structure to[br]the beam. So you could potentially do 0:52:29.510,0:52:33.320 that. So, for example, for these longer[br]periods where the kicker magnets need to 0:52:33.320,0:52:36.910 ramp up, they have something they call the[br]abort gap. So a number of bunches that are 0:52:36.910,0:52:41.350 never filled but are always kept empty. So[br]the magnets have enough time to deflect 0:52:41.350,0:52:44.460 the beam when the next bunch comes around.[br]So you could probably measure that, but 0:52:44.460,0:52:47.260 it's much easier to do it the other way[br]around. 0:52:47.260,0:52:54.570 Herald: Microphone number four, please.[br]Q: You said you had quite tight needs for 0:52:54.570,0:53:01.230 the timing clock. Is it tight enough? That[br]the speed of light was the limit with the 0:53:01.230,0:53:03.589 distances between locations or that was[br]not a concern? 0:53:03.589,0:53:08.680 Stefan: No, it is a concern. So because[br]just distributing a cable for 27 0:53:08.680,0:53:14.869 kilometers produces like just considerable[br]run times of electrical signals. All the 0:53:14.869,0:53:18.150 delays of all the cables need to be[br]measured precisely for their delay and 0:53:18.150,0:53:22.849 then calibrated out so all the experiments[br]get their clocks at the right time, 0:53:22.849,0:53:27.400 shifted, compensated for the delay time[br]that it just takes to get the signal 0:53:27.400,0:53:33.000 there.[br]Herald: And again, the interwebs. 0:53:33.000,0:53:40.210 Signal Angel: Is it too dangerous to stand[br]near the concrete cooling blocks, like 0:53:40.210,0:53:46.980 radioactive wise or, I don't know.[br]Severin: Yes. 0:53:46.980,0:53:49.090 laughter[br]Stefan: Not recommended. 0:53:49.090,0:53:54.859 Severin: Not recommended. We have a very[br]good interlock system. Also, the doors, 0:53:54.859,0:53:58.270 all the doors have switches. So basically[br]when one door is basically like opened 0:53:58.270,0:54:03.059 then basically the whole machine will be[br]shut down. So we have very critical and 0:54:03.059,0:54:11.319 safety related access system at LHC. Maybe[br]you watch Angels and Demons. This 0:54:11.319,0:54:16.760 Hollywood movie that we have, the eye[br]scanners are shown. It's a little bit. I 0:54:16.760,0:54:22.430 mean, it's Hollywood. But, we have eye[br]scanners. So iris scanners. So every time 0:54:22.430,0:54:25.780 like we want to go to the tunnel, for[br]example, then we have to let also our iris 0:54:25.780,0:54:30.000 be scanned because otherwise we will not[br]be able to go to the tunnel. So there's a 0:54:30.000,0:54:33.890 very sophisticated access system to really[br]go to the tunnel. So when there is 0:54:33.890,0:54:36.810 operation, the whole tunnel access is[br]completely blocked. 0:54:36.810,0:54:40.510 Herald: Good, microphone number one,[br]please. 0:54:40.510,0:54:47.780 Q: What is the exact reason to have each[br]of the experiments, every side. I mean, so 0:54:47.780,0:54:52.080 far apart on the LHC. I mean, on[br]opposite sides. 0:54:52.080,0:54:57.609 Severin: Um, basically, you are talking[br]about Atlas and CMS. The reason for that 0:54:57.609,0:55:02.060 is because when, these two experiments[br]were constructed, there was a little bit 0:55:02.060,0:55:10.060 of fear that particles basically interact[br]at the two experiments. So that they 0:55:10.060,0:55:13.670 really are like the most far away. We like[br]to have a very big distance from each 0:55:13.670,0:55:17.930 other. So there is no interaction between[br]them. That's why we basically put them at 0:55:17.930,0:55:20.869 point one and point five. That's the[br]reason why. 0:55:20.869,0:55:25.549 Herald: If I can see it correctly.[br]Microphone number five. 0:55:25.549,0:55:32.030 Q: Yes, hello. I've seen that you're also[br]using the CAN bus. What are you using the 0:55:32.030,0:55:37.190 CAN bus for in CERN?[br]Stefan: I know of at least one use, but it 0:55:37.190,0:55:42.539 is inside an experiment. So there are, as[br]far as I know, investigations under way to 0:55:42.539,0:55:47.610 use the CAN bus to do the actual control[br]of the detectors of one experiment. I 0:55:47.610,0:55:51.059 don't know if there is a use inside the[br]accelerator itself. So apart from the 0:55:51.059,0:55:55.790 experiments. But perhaps if you come by[br]afterwards we can find one. 0:55:55.790,0:55:59.539 Q: Thank you.[br]Herald: Microphone number one. 0:55:59.539,0:56:05.829 Q: Do you have any official data about how[br]many tons of duct taper used in 0:56:05.829,0:56:08.480 daily operations?[br]laughter 0:56:08.480,0:56:13.750 Severin: No. No.[br]Herald: What about zip ties? 0:56:13.750,0:56:21.460 Severin: Many. Yeah. Millions. Billions.[br]Herald: Okay. As far as I can see... Ah, 0:56:21.460,0:56:25.470 the intercepts again with a question.[br]Signal Angel: Do you know your monthly 0:56:25.470,0:56:27.470 power bill?[br]laughter 0:56:27.470,0:56:31.470 Severin: No, not no. No, sorry.[br]Stefan: No. But it is, I think, in fact 0:56:31.470,0:56:37.960 the contribution of France, which is the[br]main contributor in terms of energy. That 0:56:37.960,0:56:41.440 it is part of their contribution to[br]contribute the electricity bill basically 0:56:41.440,0:56:44.440 instead of paying money to CERN. That's as[br]far as I know. 0:56:44.440,0:56:48.480 Severin: Yes. And also we shut down the[br]LHC and the accelerator complex through 0:56:48.480,0:56:52.859 like the wintertime. And one of the[br]reasons for that is because electricity is 0:56:52.859,0:56:56.940 more expensive during wintertime in France[br]than in summer. 0:56:56.940,0:57:03.529 Herald: In this case, I can't see any[br]other questions. I have a maybe stupid 0:57:03.529,0:57:09.780 question. You said earlier you have to[br]focus and defocus the beam. But as we 0:57:09.780,0:57:12.529 know, you accelerated already the[br]particles. Why do we have to focus the 0:57:12.529,0:57:17.020 beam?[br]Severin: Because every time when we have a 0:57:17.020,0:57:21.910 dipole magnet, then basically we bend the[br]particle around an arc. But then we also 0:57:21.910,0:57:24.720 defocus a little bit. And also, the[br]coulomb force is still a problem because 0:57:24.720,0:57:29.770 we have equally charged particles in the[br]bunch or in the whole beam itself. So they 0:57:29.770,0:57:33.539 will by themselves basically go out of[br]each other. And if you would not focus it 0:57:33.539,0:57:36.299 again, then basically we would lose the[br]whole beam in the end. 0:57:36.299,0:57:44.630 Herald: Oh, thank you. I don't see any[br]questions. Internet? In this case thank 0:57:44.630,0:57:49.730 you very, very much, Stefan and Severin.[br]Please. With a warm applause. The Large 0:57:49.730,0:57:50.730 Hadron infrastructure talk. 0:57:50.730,0:57:51.730 Applause 0:57:51.730,0:57:52.989 subtitles created by c3subtitles.de[br]in the year 20??. Join, and help us!